Publish Ropedia Xperience-10M derived artifacts

ARTIFACT_GUIDE.md

@@ -110,7 +110,8 @@ research project.
 | [`results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md) | Documents the public multi-episode access path, selected 128-episode pilot plan, and data requirements. |
 | [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json) | Compact verified summary for the final selected-episode Qwen3-Omni diagnostic result, including split counts, held-out metrics, quality-target status, and adapter repo. |
 | [`results/omni_finetune/verified_public/`](results/omni_finetune/verified_public/) | Public-safe verified held-out result packages. These include metrics, predictions, reports, manifests, training metadata, validation summaries, and audit files, but not raw data or weights. |
-| [`results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/`](results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/) | Final verified Qwen3-Omni public package with 448 held-out predictions, 99.78% JSON validity, metrics, reports, training metadata, validation summaries, and package audit. |
+| [`results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/`](results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/) | Current verified Qwen3-Omni strict-label v3 public package with 448 held-out predictions, 100.00% JSON validity, metrics, reports, training metadata, validation summaries, package audit, and error-analysis tables. |
+| [`results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/`](results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/) | Historical Qwen3-Omni v2 strict-JSON package retained for prompt-contract and regression comparison. |
 | [`https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep`](https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep) | Public LoRA adapter weight repository for the final 128-episode Qwen3-Omni diagnostic run; raw Xperience-10M data and base Qwen weights remain excluded. |
 | [`results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md`](results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md) | Same-split 128-episode simple and neural metadata baselines for the 12 task ids, aligned to the 96/16/16 Qwen3-Omni split and explicit about raw-feature-only tasks. |
 | [`results/omni_finetune/multi_episode_128_task_baselines/summary_report.json`](results/omni_finetune/multi_episode_128_task_baselines/summary_report.json) | Machine-readable split counts, run configuration, simple metrics, neural metrics, and unsupported raw-feature markers for the aligned 128-episode baseline suite. |

PROJECT_README.md

@@ -72,9 +72,9 @@ before the multi-episode omni-model stage becomes a real held-out evaluation.
 | Dataset slice | One public Xperience-10M sample episode, 5,821 frames, 1,161 windows, and an 8,546-dimensional representation |
 | Modalities | Video, audio, depth, camera pose/SLAM, hand/body mocap, IMU, calibration, and language annotations |
 | Task suite | 12 human-readable embodied-AI task contracts with input, process, output, metrics, predictions, and case-study walkthroughs |
-| Baselines | Minimal linear/ridge/logistic heads plus compact PyTorch MLP task heads over the same chronological split |
+| Baselines | Minimal linear/ridge/logistic heads plus compact PyTorch MLP task heads over the same chronological split; companion simple/NN metadata baselines are also aligned to the selected 128-episode 96/16/16 split |
 | Research directions | Task mapping and extension probes for human modeling, 3D/4D reconstruction, egocentric interaction, and world modeling |
-| Scale-up path | A first selected-episode Qwen3-Omni LoRA diagnostic pilot has completed on the 96/16/16 split; it proves the multi-episode export/train/eval/package loop, but the weak held-out metrics make it a baseline for error analysis rather than a strong model. Cosmos 3/world-model and VLA/policy branches reuse the same split and package contract after their targets are implemented. |
+| Scale-up path | The selected-episode Qwen3-Omni LoRA final diagnostic result is verified on the 96/16/16 split; same-split simple/NN metadata baselines now cover the 12 task ids as a companion comparison. The Qwen result proves the multi-episode export/train/eval/package loop and meets the strict-JSON target, but weak action/subtask metrics make it a baseline for error analysis rather than a strong model. Cosmos3 now has two separate verified diagnostics: Nano future-window compatibility and Super base-weight Reasoner evaluation; neither is a new fine-tuned Cosmos weight release yet. |
 | Public surfaces | GitHub repo, GitHub Pages dashboard, GHCR static-site package, HF Space, HF artifact dataset, HF baseline-model repo, and HF collection |
 
 For the fastest interpretation of the current metrics, start with
@@ -109,9 +109,9 @@ This project is best read as a staged embodied-AI research study:
 | --- | --- | --- |
 | Data understanding | One public Xperience-10M sample episode is converted into 5,821 frames, 1,161 aligned windows, and an 8,546-dimensional multimodal representation. | [`PROJECT_BRIEF.md`](PROJECT_BRIEF.md), [`PROJECT_STATUS.md`](PROJECT_STATUS.md) |
 | Task suite | Twelve human-readable tasks cover action, procedure, contact, object, language, retrieval, reconstruction, order, and synchronization questions. | [`RESEARCH_TAKEAWAYS.md`](RESEARCH_TAKEAWAYS.md), [`results/episode_task_suite/summary_report.json`](results/episode_task_suite/summary_report.json) |
-| Baselines | Minimal heads and compact PyTorch MLP heads provide a first controlled comparison on the same chronological split. | [`results/episode_task_suite/neural_mlp/`](results/episode_task_suite/neural_mlp/) |
+| Baselines | Minimal heads and compact PyTorch MLP heads provide a first controlled comparison on the same chronological split; the selected 128-episode setup also has same-split simple/NN metadata baselines for JSON-supported tasks. | [`results/episode_task_suite/neural_mlp/`](results/episode_task_suite/neural_mlp/), [`results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md`](results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md) |
 | Diagnostics | Audio contribution, modality ablations, timeline overlays, object labels, and alignment stress tests show which signals are useful and which tasks remain hard. | [`results/audio_ablation/AUDIO_ABLATION_SUMMARY.md`](results/audio_ablation/AUDIO_ABLATION_SUMMARY.md), [`docs/single_episode_explorer.html`](docs/single_episode_explorer.html) |
-| Scale-up | The selected 128-episode Qwen3-Omni LoRA diagnostic pilot has a verified validation-aware held-out package: 96/16/16 selected episodes, 3,808 exported windows, 512 validation windows, 448 held-out test windows, and public-safe metrics/predictions. JSON validity is 87.50%, below the 98% target, so the next pass focuses on structured-output reliability and task-quality error analysis. | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json), [`results/omni_finetune/verified_public/`](results/omni_finetune/verified_public/) |
+| Scale-up | The selected 128-episode Qwen3-Omni LoRA diagnostic path has a final verified held-out package: 96/16/16 selected episodes, 3,808 exported windows, 512 validation windows, 448 held-out test windows, and public-safe metrics/predictions. Same-split simple/NN metadata baselines are published for the 12 task ids. Cosmos3-Nano has a verified future-window compatibility package, and Cosmos3-Super Reasoner has a verified 448-window base-weight held-out JSON-task evaluation. The comparison now supports both views: the three result layers and a model-family grouping for task heads, Qwen3-Omni LoRA, Cosmos3-Nano, and Cosmos3-Super. | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/omni_model_comparison.json`](docs/data/omni_model_comparison.json), [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json), [`results/omni_finetune/OMNI_MODEL_COMPARISON.md`](results/omni_finetune/OMNI_MODEL_COMPARISON.md), [`results/omni_finetune/verified_public/`](results/omni_finetune/verified_public/), [`results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md`](results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md) |
 
 Detailed dataset notes, reproduction checks, and generated JSON reports are
 included for readers who want to inspect the implementation, but they are
@@ -133,7 +133,7 @@ They give the current research state in one compact table:
 | Dataset context | Official Xperience-10M links, sample-vs-gated-data boundary, modality coverage, and redistribution policy are documented |
 | Evaluation protocol | Verified generated protocol for windowing, split policy, leakage controls, and per-task metrics |
 | Website and Hub pages | Public dashboard, Hugging Face Space, artifact dataset, baseline model repo, and collection use the same project framing and links |
-| Qwen3-Omni multi-episode pilot | Verified diagnostic result package exists for the selected 96/16/16 episode split; current held-out metrics are weak and below the JSON-validity quality target |
+| Qwen3-Omni multi-episode pilot | Final verified diagnostic result package exists for the selected 96/16/16 episode split; JSON validity meets the target, while action/subtask metrics remain weak |
 | Raw Xperience-10M data / full Qwen weights | Not redistributed |
 
 ## 90-Second Research Project Path
@@ -152,7 +152,7 @@ If you are reading the project cold, open these in order:
 | 8 | What research directions does this support? | [`RESEARCH_ROADMAP.md`](RESEARCH_ROADMAP.md), [`docs/data/research_directions.json`](docs/data/research_directions.json), [`docs/data/research_direction_extensions.json`](docs/data/research_direction_extensions.json) | The tasks are mapped to human modeling, 3D/4D reconstruction, egocentric interaction, and world modeling. |
 | 9 | Which foundation model comes next? | [`FOUNDATION_MODEL_PLAN.md`](FOUNDATION_MODEL_PLAN.md), [`docs/data/foundation_model_plan.json`](docs/data/foundation_model_plan.json), [`XPERIENCE_EMBODIED_FOUNDATION_MODEL_PRETRAINING.md`](XPERIENCE_EMBODIED_FOUNDATION_MODEL_PRETRAINING.md) | Qwen3-Omni is the first held-out LoRA baseline; Cosmos 3 is the first world-model branch; policy models wait for explicit action targets; Xperience-native pretraining is the full-corpus future goal. |
 | 10 | How do I reproduce it? | [`REPRODUCIBILITY.md`](REPRODUCIBILITY.md), [`notes/reproducibility_audit.md`](notes/reproducibility_audit.md) | Public commands and expected outputs are documented for the sample-episode task suite. |
-| 11 | What is still pending? | [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json), [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md) | The first held-out diagnostic pilot is verified; strong model quality remains pending because JSON validity is 87.50% and action/subtask metrics remain weak. |
+| 11 | What is still pending? | [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json), [`DATA_ACCESS_STATUS.md`](results/omni_finetune/DATA_ACCESS_STATUS.md), [`MULTI_EPISODE_ACCESS_STATUS.md`](results/omni_finetune/MULTI_EPISODE_ACCESS_STATUS.md) | The final held-out diagnostic Qwen pass is verified and JSON-validity target is met; strong action/subtask model quality remains pending. |
 
 A compact reader-path summary is available at
 [`docs/data/project_packet.json`](docs/data/project_packet.json).
@@ -481,8 +481,8 @@ python scripts/train_all_modalities_model.py --workspace /path/to/workspace
 
 This repo includes a first Qwen3-Omni fine-tuning path over Xperience-10M. The
 repository separates public-sample evidence from multi-episode fine-tuning
-artifacts. The validation-aware selected-episode held-out package is now verified as a
-diagnostic pilot, not a strong final model.
+artifacts. The selected-episode held-out package is now verified as a
+diagnostic result, not a strong final action/subtask model.
 The useful distinction is:
 
 - direct Qwen3-Omni inputs: RGB/fisheye video, embedded MP4 audio, and language
@@ -505,6 +505,15 @@ for public README, website, or Hugging Face updates only after the validator
 passes and `scripts/omni/package_verified_omni_result.py` creates a
 public-safe derived-artifact package. The current verified package is listed in
 [`docs/data/omni_finetune_verified_result.json`](docs/data/omni_finetune_verified_result.json).
+The current cross-version comparison is generated at
+[`docs/data/omni_model_comparison.json`](docs/data/omni_model_comparison.json)
+and [`results/omni_finetune/OMNI_MODEL_COMPARISON.md`](results/omni_finetune/OMNI_MODEL_COMPARISON.md);
+it separates the single-episode task suite, 128-episode aligned simple/NN
+baselines, and verified Qwen3/Cosmos model-branch packages. The same generated
+files also include `model_groups`: a model-first view that pairs 1-episode and
+128-episode entries for the same family. Use that section when comparing task
+heads against task heads, Qwen3-Omni smoke/LoRA against Qwen3-Omni LoRA, or
+Cosmos3-Nano compatibility against future Cosmos weight releases.
 
 ### Sample Count Decision
 
@@ -544,13 +553,18 @@ Current status in this repo:
 - gated_metadata_audit: 12,102 complete visible episodes across 802 complete sessions
 - selected_episode_plan: 128 source-balanced episodes, 96/16/16 train/val/test
 - selected_download_size: 277.71 GiB excluding `visualization.rrd`
-- verified_validation_aware_diagnostic_package: true
+- verified_final_diagnostic_package: true
 - selected_split: 96 train / 16 validation / 16 held-out test episodes
 - exported_windows: 2,848 train / 512 validation / 448 test
 - validation_samples_used: 512
 - held_out_eval: 448 test windows from 14 exported test episodes
-- train_loss / val_loss: 0.4130 / 0.0331
-- current_quality_target: JSON validity 87.50%, below the 98% target
+- final_train_loss / final_val_loss: 0.0277 / 0.0278
+- current_quality_target: strict-label JSON validity 100.00%, meeting the 98% target; action/subtask quality remains weak
+- qwen3_lora_adapter_repo: https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep
+- 128_aligned_baselines: 12 task ids, 8 simple metadata/text baselines, 6 neural metadata/text baselines
+- cosmos3_nano_branch: verified Cosmos3-Nano future-window compatibility package, 378 held-out future-window predictions from 14 test episodes
+- cosmos3_super_branch: verified Cosmos3-Super Reasoner base-weight JSON-task evaluation, 448 held-out predictions from 14 test episodes; JSON validity 51.12%, action macro-F1 0.0008, contact accuracy 32.14%, transition accuracy 36.83%
+- cosmos3_super_training_readiness: runtime/GPU probe passes for Diffusers on 8 CUDA devices, but true fine-tuning is blocked until a Cosmos3-specific diffusion/action target packer and supervised loss are implemented; no Cosmos weights have been updated
 - gated dataset: available for selected multi-episode data preparation
 - source_discovery: `results/omni_finetune/source_discovery.json`
 - data_status: `results/omni_finetune/DATA_ACCESS_STATUS.md`
@@ -678,21 +692,10 @@ For hardware setups that can run multiple eval workers, the Qwen evaluator also
 supports deterministic sample shards:
 
 ```bash
-python scripts/omni/eval_qwen3_omni_lora.py \
-  --dataset-jsonl results/omni_finetune/xperience10m_qwen3_omni_128ep_fullsplit_fast8gpu_dataset/dataset.jsonl \
-  --adapter-dir checkpoints/<train_run_id>/adapter_lora \
-  --run-id <eval_shard_0> \
-  --eval-split test \
-  --sample-offset 0 \
-  --sample-stride 4
-
-python scripts/omni/merge_qwen3_omni_eval_shards.py \
-  --dataset-jsonl results/omni_finetune/xperience10m_qwen3_omni_128ep_fullsplit_fast8gpu_dataset/dataset.jsonl \
-  --output-dir results/omni_finetune/<merged_eval_run_id> \
-  --shard-dir results/omni_finetune/<eval_shard_0> \
-  --shard-dir results/omni_finetune/<eval_shard_1> \
-  --shard-dir results/omni_finetune/<eval_shard_2> \
-  --shard-dir results/omni_finetune/<eval_shard_3>
+CUDA_DEVICE_GROUPS="0,1 2,3 4,5 6,7" \
+SHARDS=4 \
+RUN_ID=<merged_eval_run_id> \
+scripts/omni/run_qwen3_omni_lora_eval_sharded.sh
 ```
 
 Only the merged eval directory should be validated and reported publicly,
@@ -716,12 +719,19 @@ windows without depending on Qwen chat-message records.
 The public-safe verified package intentionally excludes raw data, base Qwen
 weights, LoRA weights, and full checkpoints. Adapter upload is a separate step:
 use it only when the intended adapter directory is present and the model card
-clearly distinguishes older smoke weights from the selected-episode diagnostic
-or validation-aware run.
+clearly distinguishes older smoke weights from the final selected-episode
+diagnostic run.
+
+Keep weight-bearing repositories model-specific: the final 128-episode
+Qwen3-Omni adapter belongs in `cy0307/ropedia-qwen3-omni-lora-128ep`, older
+Qwen smoke material remains historical. The current Cosmos3-Nano and
+Cosmos3-Super outputs are artifacts-only diagnostics; Cosmos3 should get its
+own model repo only after real Cosmos adapter or fine-tuned weights exist.
+Metrics, predictions, audits, and reports stay in the artifact dataset.
 
 ```bash
 python3 scripts/omni/upload_qwen3_omni_lora_to_hf.py \
-  --repo-id cy0307/ropedia-qwen3-omni-lora-smoke \
+  --repo-id cy0307/ropedia-qwen3-omni-lora-128ep \
   --source-dir /path/to/adapter_upload_package \
   --message "Upload Xperience-10M Qwen3-Omni LoRA pilot"
 ```

PROJECT_STATUS.md

@@ -31,7 +31,7 @@ scale-up readiness; it is not presented as final full-dataset model quality.
 | Public package policy | Verified | `DATA_NOTICE.md`, `REPRODUCIBILITY.md` | Raw Xperience-10M data, private gated files, large archives, credentials, and full Qwen weights are not redistributed. |
 | Reproducibility | Verified for the public sample | `REPRODUCIBILITY.md`, `docs/data/reproducibility_matrix.json`, `notes/reproducibility_audit.md` | The public sample workflow has explicit commands, expected outputs, and exact-match reproduction evidence. |
 | 128-episode aligned baselines | Verified companion result | `results/omni_finetune/multi_episode_128_task_baselines/BASELINE_ALIGNMENT_REPORT.md`, `results/omni_finetune/multi_episode_128_task_baselines/summary_report.json`, `scripts/omni/run_128_task_baselines.py` | The earlier simple and neural baseline framing is aligned to the same selected 96/16/16 episode split used by the Qwen3-Omni pilot. JSON-supported tasks have metadata/text simple and neural MLP metrics; raw-feature-only tasks are explicitly marked unsupported until 128-run sensor feature blocks are available. |
-| Qwen3-Omni fine-tuning | Final verified diagnostic held-out result; JSON target met | `docs/data/omni_finetune_verified_result.json`, `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/`, `scripts/omni/package_verified_omni_result.py`, `scripts/omni/audit_verified_omni_package.py`, `scripts/omni/analyze_qwen3_omni_errors.py` | The selected 96/16/16 episode split produced a final public-safe held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs, and validation/audit summaries. JSON validity is 99.78%, meeting the 98% target; transition accuracy is 97.10%, contact accuracy is 71.88%, object micro-F1 is 30.16%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim. |
+| Qwen3-Omni fine-tuning | Final verified diagnostic held-out result; JSON target met | `docs/data/omni_finetune_verified_result.json`, `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/`, `scripts/omni/package_verified_omni_result.py`, `scripts/omni/audit_verified_omni_package.py`, `scripts/omni/analyze_qwen3_omni_errors.py` | The selected 96/16/16 episode split produced a current public-safe strict-label v3 held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs, validation/audit summaries, and the reused public LoRA adapter. JSON validity is 100.00%, meeting the 98% target; transition accuracy is 97.32%, contact accuracy is 72.10%, object micro-F1 is 30.69%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim. |
 | Raw Xperience-10M redistribution | Not included | `DATA_NOTICE.md`, `docs/data/publication_audit.json` | Raw MP4, HDF5, RRD files, private gated data, and full Qwen weights are intentionally excluded. |
 
 ## Fast Research Route
@@ -69,9 +69,10 @@ scale-up readiness; it is not presented as final full-dataset model quality.
   current results use one public sample episode.
 - Public-facing fine-tuning results should come from the verified result
   package, not from live process logs or setup-only artifacts.
-- The final Qwen3-Omni held-out package verifies the pipeline and meets the
-  strict-JSON target, but not strong action/subtask model quality: JSON validity
-  is 99.78%, action macro-F1 is 0.0024, and subtask accuracy is 0.0022.
+- The final Qwen3-Omni strict-label v3 held-out package verifies the pipeline
+  and meets the strict-JSON target, but not strong action/subtask model quality:
+  JSON validity is 100.00%, action macro-F1 is 0.0022, and subtask accuracy is
+  0.0022.
 - The 128-episode aligned simple/NN baselines use metadata/text features from
   the derived Qwen JSONL export; they align the split and task ids but do not
   replace raw-modality baselines for trajectory, retrieval, reconstruction, or

README.md

@@ -21,7 +21,7 @@ configs:
 This dataset repository stores small derived artifacts for the Ropedia
 Xperience-10M task-suite project: metrics, predictions, manifests, reports,
 figures, website JSON, public-safe Qwen3-Omni diagnostic outputs, and the
-Cosmos3-Nano future-window compatibility package.
+Cosmos3-Nano plus Cosmos3-Super diagnostic packages.
 
 ## What To Open First
 
@@ -48,9 +48,9 @@ The implemented public-sample task suite uses one public Xperience-10M sample
 episode. The selected 128-episode Qwen3-Omni final diagnostic result uses a
 gated local dataset copy and publishes only public-safe metrics, predictions,
 manifests, reports, and audits. The public LoRA adapter weights are published
-separately at `cy0307/ropedia-qwen3-omni-lora-128ep`. Cosmos3-Nano is currently
-published here as an artifacts-only future-window compatibility result; create a
-separate Cosmos model repository only after real Cosmos adapter or fine-tuned
+separately at `cy0307/ropedia-qwen3-omni-lora-128ep`. Cosmos3-Nano and
+Cosmos3-Super are currently published here as artifacts-only diagnostics; create
+a separate Cosmos model repository only after real Cosmos adapter or fine-tuned
 weights exist.
 
 ## Derived Artifacts
@@ -61,15 +61,19 @@ This bundle includes derived artifacts such as:
   MLP result directories.
 - Audio-ablation summaries and generated chart assets.
 - Public website JSON and figure manifests.
-- A final verified Qwen3-Omni diagnostic held-out package for the selected
-  96/16/16 episode split, with 99.78% JSON validity and weak action/subtask
-  quality documented as the next error-analysis target.
-- A verified Cosmos3-Nano future-window compatibility package for the same
-  selected split family.
+- A current verified Qwen3-Omni strict-label v3 held-out package for the
+  selected 96/16/16 episode split, with 100.00% JSON validity and weak
+  action/subtask quality documented as the next error-analysis target.
+- Historical Qwen3-Omni packages, including the earlier v2 strict-JSON
+  diagnostic, for regression and prompt-contract comparison.
+- Verified Cosmos3-Nano future-window compatibility and Cosmos3-Super
+  base-weight/readiness packages for the same selected split family; these are
+  artifacts only, not new Cosmos fine-tuned weight releases.
 - 128-episode same-split simple/NN metadata baselines for the same 12 task ids,
   with unsupported markers where raw 128 sensor feature blocks are still needed.
 - A model-family grouped comparison that pairs 1-episode and 128-episode entries
-  for task heads, Qwen3-Omni LoRA, and Cosmos3-Nano without mixing target types.
+  for task heads, Qwen3-Omni LoRA, Cosmos3-Nano, and Cosmos3-Super without
+  mixing target types.
 
 ## Related Hub Repositories
 

RESEARCH_ROADMAP.md

@@ -95,10 +95,10 @@ Evidence to inspect:
 This stage uses Qwen3-Omni as the multimodal backbone and trains lightweight
 LoRA adapters. The final held-out diagnostic package now exists. It proves the
 export, training, evaluation, validation, public-safe packaging, and adapter
-publication loop. JSON validity is 99.78%, transition accuracy is 97.10%, and
-contact accuracy is 71.88%, but action macro-F1 is 0.0024 and subtask accuracy
-is 0.0022. Treat it as a baseline and error-analysis starting point, not as a
-strong action/subtask model.
+publication loop. The current strict-label v3 evaluation reaches 100.00% JSON
+validity, 97.32% transition accuracy, and 72.10% contact accuracy, but action
+macro-F1 is 0.0022 and subtask accuracy is 0.0022. Treat it as a baseline and
+error-analysis starting point, not as a strong action/subtask model.
 
 Expected outputs:
 

data/artifact_index.json

@@ -1,19 +1,19 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-06T23:27:35+00:00",
+  "generated_at_utc": "2026-06-07T15:47:31+00:00",
   "status": "pass",
-  "artifact_count": 118,
+  "artifact_count": 129,
   "missing": [],
   "by_kind": {
     "project_path": 14,
     "scaleup_contract": 7,
-    "scaleup_status": 16,
+    "scaleup_status": 20,
     "publication_workflow": 5,
     "project_scope": 1,
     "source_alignment": 5,
     "evaluation_protocol": 3,
     "result_interpretation": 5,
-    "metrics_source": 14,
+    "metrics_source": 18,
     "website_data": 3,
     "visual_evidence": 7,
     "quality_gate": 12,
@@ -31,8 +31,8 @@
     "generated_figure_assets": 1,
     "citation": 1,
     "license": 1,
-    "verified_public_package": 4,
-    "publication_audit": 3
+    "verified_public_package": 6,
+    "publication_audit": 4
   },
   "artifacts": [
     {
@@ -65,8 +65,8 @@
       "surface": "repo_hf",
       "shows": "Gives a compact current-state table for first-pass readers.",
       "exists": true,
-      "bytes": 9845,
-      "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+      "bytes": 9926,
+      "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
     },
     {
       "id": "project_status_json",
@@ -76,8 +76,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable copy of the current project status for website and HF mirrors.",
       "exists": true,
-      "bytes": 15049,
-      "sha256": "23873ed59f3a38f46e45b15a5965afbb1365d49eb359bd5089a4ba6bda990d3c"
+      "bytes": 16455,
+      "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
     },
     {
       "id": "research_roadmap",
@@ -87,8 +87,8 @@
       "surface": "repo_hf",
       "shows": "Defines the path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
       "exists": true,
-      "bytes": 12194,
-      "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+      "bytes": 12233,
+      "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
     },
     {
       "id": "research_roadmap_json",
@@ -142,8 +142,8 @@
       "surface": "repo_hf",
       "shows": "Stores the implemented Qwen3-Omni LoRA contract and planned Cosmos-style world-model and VLA/policy branch contracts.",
       "exists": true,
-      "file_count": 3,
-      "bytes": 9203
+      "file_count": 4,
+      "bytes": 12613
     },
     {
       "id": "omni_backbone_registry_validator",
@@ -197,8 +197,8 @@
       "surface": "repo_hf",
       "shows": "Computes public-safe held-out error-analysis tables by episode, action family, train-seen status, required-modality state, and object category.",
       "exists": true,
-      "bytes": 15676,
-      "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+      "bytes": 16957,
+      "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
     },
     {
       "id": "multi_episode_128_baseline_script",
@@ -219,8 +219,8 @@
       "surface": "repo_hf",
       "shows": "Builds the upload-ready Hugging Face adapter folder from a verified Qwen3 LoRA result summary and adapter directory.",
       "exists": true,
-      "bytes": 9843,
-      "sha256": "636132a7d299db4d874ec797e34acd7e37eea69994c2d39afaafaec6587169a0"
+      "bytes": 10710,
+      "sha256": "c5f33f7030e6861ccab465b71d4c1b4209b8fbbcebdca882aa5c7e8d70e76b37"
     },
     {
       "id": "additional_development_directions",
@@ -274,8 +274,8 @@
       "surface": "website_hf",
       "shows": "Gives a short project path with scope status and public surfaces.",
       "exists": true,
-      "bytes": 7943,
-      "sha256": "ffd5da5fd2c2dc82fa1beb74335a51a33317923b3e7ee4864e2b5031082b0a42"
+      "bytes": 8005,
+      "sha256": "2258fecb80850c745e60cb28733869c49a5182879d9d0461b666a5575e3c1610"
     },
     {
       "id": "artifact_guide",
@@ -285,8 +285,8 @@
       "surface": "repo_hf",
       "shows": "Gives the human-readable map from project scope to data, tasks, platform mirrors, and scale-up status.",
       "exists": true,
-      "bytes": 17508,
-      "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+      "bytes": 17930,
+      "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
     },
     {
       "id": "official_dataset_card_alignment",
@@ -686,7 +686,7 @@
       "volatile": true,
       "shows": "Records the last live GitHub/HF URL verification after upload.",
       "exists": true,
-      "bytes": 68749,
+      "bytes": 74880,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -697,8 +697,8 @@
       "surface": "repo",
       "shows": "Fetches the published GitHub/HF URLs and compares live hashes and public-card markers against the release assets.",
       "exists": true,
-      "bytes": 36847,
-      "sha256": "07fd059a9ff8c13b073f349c79f1f7d3abe839559cf0809e291f6ea9bbad21e8"
+      "bytes": 36201,
+      "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
     },
     {
       "id": "reproducibility_contract",
@@ -719,8 +719,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable reproduction steps with expected artifacts and public boundaries.",
       "exists": true,
-      "bytes": 5280,
-      "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+      "bytes": 5294,
+      "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
     },
     {
       "id": "artifact_index_builder",
@@ -754,7 +754,7 @@
       "volatile": true,
       "shows": "Separates setup paths from completed held-out-episode results.",
       "exists": true,
-      "bytes": 21234,
+      "bytes": 21251,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -766,7 +766,7 @@
       "volatile": true,
       "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
       "exists": true,
-      "bytes": 235815,
+      "bytes": 410374,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -965,8 +965,8 @@
       "surface": "repo_hf",
       "shows": "Documents the final 128-episode LoRA adapter upload path, target model repo, package builder, and forbidden files.",
       "exists": true,
-      "bytes": 1875,
-      "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+      "bytes": 1886,
+      "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
     },
     {
       "id": "multi_episode_access_status",
@@ -1031,8 +1031,8 @@
       "surface": "repo_hf",
       "shows": "Reader-facing comparison of the single-episode task suite, 128-episode aligned baselines, Qwen3-Omni packages, and Cosmos3 future-window branch.",
       "exists": true,
-      "bytes": 3110,
-      "sha256": "11c22b7ac1e16fd8db86eb7c6fc33cf28fee97a38098f1606a35daee113dc72b"
+      "bytes": 9231,
+      "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
     },
     {
       "id": "omni_model_comparison_json",
@@ -1042,8 +1042,8 @@
       "surface": "repo_hf",
       "shows": "Machine-readable comparison of the current result versions, per-task aligned baselines, verified Qwen3 packages, and Cosmos3 package.",
       "exists": true,
-      "bytes": 21433,
-      "sha256": "b539a489a8974ecec90dda312471be54f466b81bef9d1ebc99d08155f8c21c94"
+      "bytes": 48296,
+      "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
     },
     {
       "id": "cosmos3_nano_verified_summary",
@@ -1144,6 +1144,61 @@
       "bytes": 1099,
       "sha256": "f11ccb167908d4f5bfb49c0be0b4bc6c9254901462aa52ae98a2a98e8af16558"
     },
+    {
+      "id": "verified_public_package_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified public package: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "kind": "verified_public_package",
+      "surface": "repo_hf",
+      "shows": "Public-safe verified package for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607 (cosmos3_super_reasoner, status=verified).",
+      "exists": true,
+      "file_count": 21,
+      "bytes": 5151738
+    },
+    {
+      "id": "verified_public_summary_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified summary: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Machine-readable verified summary for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 5693,
+      "sha256": "6e263015a73dbcd8d9dd2fa036348d1a20c3e4aecfd1f6bf43d1b9f923320b91"
+    },
+    {
+      "id": "verified_public_public_result_summary_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified public result summary: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/PUBLIC_RESULT_SUMMARY.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Public result summary for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 979,
+      "sha256": "1c17ac8b08e522abaf606a32982086295c577c81f94a3495ea2b9b6a50f83d74"
+    },
+    {
+      "id": "verified_public_run_report_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified run report: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/eval/RUN_REPORT.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Run report for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 549,
+      "sha256": "9a3edf661c75956335137f98fa4cd40071bcc0a66e7553bb614375f2e2e2ac22"
+    },
+    {
+      "id": "verified_public_metrics_JSON_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified metrics JSON: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/eval/metrics.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Metrics json for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 48156,
+      "sha256": "138b9de0ba2cda7b273c47a59656f5abc85c435a64108d9932bc3611143e5d8e"
+    },
     {
       "id": "verified_public_package_xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval",
       "title": "Verified public package: Qwen3-Omni LoRA",
@@ -1284,8 +1339,8 @@
       "surface": "repo_hf",
       "shows": "Public-safe verified package for xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full (qwen3_omni_lora, status=verified).",
       "exists": true,
-      "file_count": 16,
-      "bytes": 4898687
+      "file_count": 23,
+      "bytes": 4923819
     },
     {
       "id": "verified_public_summary_xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full",
@@ -1341,6 +1396,72 @@
       "exists": true,
       "bytes": 623,
       "sha256": "d7264cfb34e48b5c41c89444ea9cd1314b8f4d0bcc0224debbbe5ea512450197"
+    },
+    {
+      "id": "verified_public_package_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified public package: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "kind": "verified_public_package",
+      "surface": "repo_hf",
+      "shows": "Public-safe verified package for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full (qwen3_omni_lora, status=verified).",
+      "exists": true,
+      "file_count": 23,
+      "bytes": 4879339
+    },
+    {
+      "id": "verified_public_summary_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified summary: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Machine-readable verified summary for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 6208,
+      "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+    },
+    {
+      "id": "verified_public_public_result_summary_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified public result summary: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Public result summary for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 1002,
+      "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+    },
+    {
+      "id": "verified_public_run_report_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified run report: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Run report for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 580,
+      "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+    },
+    {
+      "id": "verified_public_metrics_JSON_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified metrics JSON: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Metrics json for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 47129,
+      "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+    },
+    {
+      "id": "verified_public_package_audit_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified package audit: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+      "kind": "publication_audit",
+      "surface": "repo_hf",
+      "shows": "Package audit for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 634,
+      "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
     }
   ]
 }

data/mirror_parity.json

@@ -1,9 +1,9 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:41:47+00:00",
+  "generated_at_utc": "2026-06-07T15:49:31+00:00",
   "hf_root": "hf_publish",
   "summary": {
-    "group_count": 197,
+    "group_count": 234,
     "failure_count": 0,
     "failures_by_surface": {}
   },
@@ -102,27 +102,27 @@
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
-        "bytes": 60162,
-        "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+        "bytes": 67401,
+        "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         }
       },
       "failures": []
@@ -289,26 +289,26 @@
         "path": "repo:docs/data/live_publication_status.json",
         "exists": true,
         "bytes": 74880,
-        "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+        "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         },
         "hf_model": {
           "path": "hf_model:metrics/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         }
       },
       "failures": []
@@ -350,27 +350,27 @@
       "local": {
         "path": "repo:docs/data/omni_finetune_verified_result.json",
         "exists": true,
-        "bytes": 3483,
-        "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+        "bytes": 3628,
+        "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         },
         "hf_model": {
           "path": "hf_model:metrics/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         }
       },
       "failures": []
@@ -381,27 +381,27 @@
       "local": {
         "path": "repo:docs/data/omni_model_comparison.json",
         "exists": true,
-        "bytes": 42015,
-        "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+        "bytes": 48296,
+        "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         },
         "hf_model": {
           "path": "hf_model:metrics/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         }
       },
       "failures": []
@@ -505,27 +505,27 @@
       "local": {
         "path": "repo:docs/data/project_status.json",
         "exists": true,
-        "bytes": 16400,
-        "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+        "bytes": 16455,
+        "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         }
       },
       "failures": []
@@ -537,26 +537,26 @@
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
         "bytes": 7212,
-        "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+        "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         }
       },
       "failures": []
@@ -660,27 +660,27 @@
       "local": {
         "path": "repo:docs/data/reproducibility_matrix.json",
         "exists": true,
-        "bytes": 5280,
-        "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+        "bytes": 5294,
+        "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         },
         "hf_model": {
           "path": "hf_model:metrics/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         }
       },
       "failures": []
@@ -846,27 +846,27 @@
       "local": {
         "path": "repo:docs/data/scope_claims_audit.json",
         "exists": true,
-        "bytes": 21234,
-        "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+        "bytes": 21251,
+        "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         },
         "hf_model": {
           "path": "hf_model:metrics/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         }
       },
       "failures": []
@@ -971,26 +971,26 @@
         "path": "repo:docs/data/task_surface_integrity.json",
         "exists": true,
         "bytes": 45779,
-        "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+        "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         }
       },
       "failures": []
@@ -1033,26 +1033,26 @@
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
         "bytes": 15375,
-        "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+        "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         }
       },
       "failures": []
@@ -1760,21 +1760,21 @@
       "local": {
         "path": "repo:scripts/omni/analyze_qwen3_omni_errors.py",
         "exists": true,
-        "bytes": 15676,
-        "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+        "bytes": 16957,
+        "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/analyze_qwen3_omni_errors.py",
           "exists": true,
-          "bytes": 15676,
-          "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+          "bytes": 16957,
+          "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/analyze_qwen3_omni_errors.py",
           "exists": true,
-          "bytes": 15676,
-          "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+          "bytes": 16957,
+          "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
         }
       },
       "failures": []
@@ -1785,21 +1785,21 @@
       "local": {
         "path": "repo:scripts/omni/build_omni_model_comparison.py",
         "exists": true,
-        "bytes": 27997,
-        "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+        "bytes": 30236,
+        "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/build_omni_model_comparison.py",
           "exists": true,
-          "bytes": 27997,
-          "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+          "bytes": 30236,
+          "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/build_omni_model_comparison.py",
           "exists": true,
-          "bytes": 27997,
-          "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+          "bytes": 30236,
+          "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
         }
       },
       "failures": []
@@ -2160,21 +2160,21 @@
       "local": {
         "path": "repo:scripts/verify_live_publication.py",
         "exists": true,
-        "bytes": 36237,
-        "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+        "bytes": 36201,
+        "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 36237,
-          "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+          "bytes": 36201,
+          "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
         },
         "hf_model": {
           "path": "hf_model:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 36237,
-          "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+          "bytes": 36201,
+          "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
         }
       },
       "failures": []
@@ -2410,21 +2410,21 @@
       "local": {
         "path": "repo:docs/index.html",
         "exists": true,
-        "bytes": 180715,
-        "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+        "bytes": 180727,
+        "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
-          "bytes": 180715,
-          "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+          "bytes": 180727,
+          "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
-          "bytes": 180715,
-          "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+          "bytes": 180727,
+          "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
         }
       },
       "failures": []
@@ -2665,27 +2665,27 @@
       "local": {
         "path": "repo:results/omni_finetune/HF_UPLOAD.md",
         "exists": true,
-        "bytes": 1875,
-        "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+        "bytes": 1886,
+        "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         }
       },
       "failures": []
@@ -2696,27 +2696,27 @@
       "local": {
         "path": "repo:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
         "exists": true,
-        "bytes": 8275,
-        "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+        "bytes": 9231,
+        "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         }
       },
       "failures": []
@@ -3279,6 +3279,223 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3361,
+        "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1225,
+        "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 1994,
+        "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 21118,
+        "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 393,
+        "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1618,
+        "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 522,
+        "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/dataset/dataset_manifest.json",
       "status": "pass",
@@ -4860,6 +5077,223 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3348,
+        "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1223,
+        "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 2044,
+        "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 15918,
+        "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 395,
+        "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1663,
+        "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 541,
+        "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/dataset/dataset_manifest.json",
       "status": "pass",
@@ -5325,6 +5759,719 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+        "exists": true,
+        "bytes": 1002,
+        "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3352,
+        "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1260,
+        "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 1967,
+        "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 15657,
+        "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 366,
+        "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1655,
+        "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 521,
+        "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+        "exists": true,
+        "bytes": 252195,
+        "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+        "exists": true,
+        "bytes": 753255,
+        "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+        "exists": true,
+        "bytes": 580,
+        "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+        "exists": true,
+        "bytes": 2948548,
+        "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+        "exists": true,
+        "bytes": 47129,
+        "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+        "exists": true,
+        "bytes": 46482,
+        "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+        "exists": true,
+        "bytes": 222559,
+        "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+        "exists": true,
+        "bytes": 550610,
+        "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+        "exists": true,
+        "bytes": 634,
+        "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+        "exists": true,
+        "bytes": 508,
+        "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+        "exists": true,
+        "bytes": 16360,
+        "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+        "exists": true,
+        "bytes": 2429,
+        "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+        "exists": true,
+        "bytes": 3354,
+        "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+        "exists": true,
+        "bytes": 2708,
+        "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+        "exists": true,
+        "bytes": 6208,
+        "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/single_episode_diagnostics/README.md",
       "status": "pass",
@@ -5610,27 +6757,27 @@
       "local": {
         "path": "repo:ARTIFACT_GUIDE.md",
         "exists": true,
-        "bytes": 17508,
-        "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+        "bytes": 17930,
+        "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         },
         "hf_model": {
           "path": "hf_model:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         }
       },
       "failures": []
@@ -5889,27 +7036,27 @@
       "local": {
         "path": "repo:RESEARCH_ROADMAP.md",
         "exists": true,
-        "bytes": 12194,
-        "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+        "bytes": 12233,
+        "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         }
       },
       "failures": []
@@ -5920,27 +7067,27 @@
       "local": {
         "path": "repo:PROJECT_STATUS.md",
         "exists": true,
-        "bytes": 9845,
-        "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+        "bytes": 9926,
+        "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         }
       },
       "failures": []

data/omni_finetune_verified_result.json

@@ -48,28 +48,29 @@
       }
     ],
     "loss": "answer-token cross entropy over supervised JSON tokens",
-    "note": "This final Qwen3-Omni LoRA pass reused the selected 96/16/16 episode setup, trained on all exported train windows with validation monitoring, and preserved the held-out test split for final evaluation."
+    "note": "This current Qwen3-Omni LoRA result reuses the selected 96/16/16 episode setup and the v2 trained adapter, then applies the stricter label-contract prompt for held-out evaluation."
   },
   "evaluation": {
     "split": "test",
     "num_samples": 448,
     "held_out_episode_count": 14,
-    "json_validity_rate": 0.9977678571428571,
-    "action_macro_f1": 0.0024331644885523347,
+    "json_validity_rate": 1.0,
+    "action_macro_f1": 0.0021983997167007384,
     "subtask_accuracy": 0.002232142857142857,
-    "transition_accuracy": 0.9709821428571429,
-    "next_action_accuracy": 0.029017857142857144,
-    "contact_accuracy": 0.71875,
-    "object_micro_f1": 0.30160427807486634,
+    "transition_accuracy": 0.9732142857142857,
+    "next_action_accuracy": 0.03125,
+    "contact_accuracy": 0.7209821428571429,
+    "object_micro_f1": 0.30688228657389993,
     "quality_target": {
       "json_validity_rate": 0.98,
       "status": "met"
     },
-    "previous_validation_aware_json_validity_rate": 0.875
+    "previous_validation_aware_json_validity_rate": 0.875,
+    "previous_structured_json_v2_json_validity_rate": 0.9977678571428571
   },
-  "interpretation": "This is the final verified two-epoch Qwen3-Omni LoRA diagnostic result for the selected 128-episode setup. It meets the 98% JSON-validity target and improves transition, contact, and object metrics over the earlier validation-aware pilot, but action and subtask classification remain weak on held-out episodes, so this is still a baseline-quality diagnostic model rather than a strong Xperience-10M action recognizer.",
+  "interpretation": "This is the current verified Qwen3-Omni LoRA diagnostic result for the selected 128-episode setup. It reuses the same trained LoRA adapter as v2 but tightens the prompt-side label contract at evaluation time, reaching 100% JSON validity and small gains in transition, contact, next-action exact accuracy, and object micro-F1. Action and subtask classification remain weak on held-out episodes, so this is still a baseline-quality diagnostic model rather than a strong Xperience-10M action recognizer.",
   "public_package": {
-    "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full",
+    "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
     "audit_status": "pass",
     "contains_raw_xperience10m_data": false,
     "contains_qwen_base_weights": false,
@@ -77,9 +78,9 @@
     "adapter_weights_repo": "cy0307/ropedia-qwen3-omni-lora-128ep"
   },
   "required_next_steps": [
-    "Verify the public Hugging Face LoRA adapter repository hashes after publication.",
-    "Publish the final verified package and refreshed comparison tables to all public mirrors, then run live publication verification.",
-    "Use the full-eval predictions for error analysis focused on action/subtask confusions and unseen-label behavior.",
-    "Keep the same verified package contract for the Cosmos3 world-model branch and any future VLA/policy branches."
+    "Use the v3 strict-label predictions for action/subtask error analysis and unseen-label debugging.",
+    "Keep the existing Qwen LoRA adapter repository as the weight-bearing artifact; v3 is an evaluation/package refresh over the same adapter, not new weights.",
+    "Implement the Cosmos3-Super diffusion/action target packer and supervised loss before claiming Cosmos3 fine-tuning.",
+    "Use sharded Qwen eval for future long held-out passes to improve GPU utilization."
   ]
 }

data/omni_model_comparison.json

@@ -1,14 +1,14 @@
 {
   "title": "Ropedia Xperience-10M Current Result Versions and Model Groups",
-  "generated_at_utc": "2026-06-07T07:49:32+00:00",
+  "generated_at_utc": "2026-06-07T15:34:51+00:00",
   "status": "pass",
   "version_count": 3,
-  "model_group_count": 3,
-  "comparison_rule": "Compare only rows with the same scope and target. Single-episode raw-feature metrics, 128-episode metadata baselines, Qwen3 structured JSON metrics, and Cosmos3 future-window metrics answer different questions.",
+  "model_group_count": 4,
+  "comparison_rule": "Compare only rows with the same scope and target. Single-episode raw-feature metrics, 128-episode metadata baselines, Qwen3 structured JSON metrics, and the two Cosmos3 targets answer different questions: Nano future-window retrieval versus Super structured JSON Reasoner evaluation.",
   "version_reading_notes": [
     "Version 1 is the public-sample 12-task harness with minimal and neural heads.",
     "Version 2 is the selected 128-episode same-split simple/NN baseline alignment.",
-    "Version 3 is the verified model-branch layer: the current final Qwen3-Omni LoRA package is the JSON-task diagnostic result, while Cosmos3-Nano is a future-window compatibility result rather than a full Cosmos diffusion fine-tune."
+    "Version 3 is the verified model-branch layer: the current final Qwen3-Omni LoRA package is the JSON-task diagnostic result, Cosmos3-Nano is a future-window compatibility result, and Cosmos3-Super Reasoner is a base-weight JSON-task evaluation rather than a new fine-tuned weight release."
   ],
   "versions": [
     {
@@ -313,13 +313,16 @@
       "source": "results/omni_finetune/verified_public/",
       "split": "episode/session held-out split; exact task target depends on backbone contract",
       "counts": {
-        "verified_branch_count": 4,
-        "qwen3_verified_package_count": 3,
-        "cosmos3_verified_package_count": 1
+        "verified_branch_count": 6,
+        "qwen3_verified_package_count": 4,
+        "cosmos3_verified_package_count": 2,
+        "cosmos3_nano_verified_package_count": 1,
+        "cosmos3_super_verified_package_count": 1
       },
       "models": [
         "Qwen3-Omni LoRA",
-        "Cosmos3-Nano future-window compatibility branch"
+        "Cosmos3-Nano future-window compatibility branch",
+        "Cosmos3-Super Reasoner base-weight evaluation"
       ],
       "branches": [
         {
@@ -367,6 +370,45 @@
           "is_current": true,
           "weights_repository": "planned separate Cosmos3 model repo after a real Cosmos diffusion/LoRA fine-tune exists; current result remains artifacts-only"
         },
+        {
+          "id": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+          "title": "Cosmos3-Super Reasoner",
+          "status": "verified",
+          "backbone": "cosmos3_super_reasoner",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "zero_shot_structured_episode_understanding_json_qa_via_vllm_reasoner",
+          "source": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_cosmos3_super_reasoner_base_vllm_8gpu_20260607",
+          "eval_run_id": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 0.5111607142857143,
+            "action_macro_f1": 0.0008284021201089245,
+            "subtask_accuracy": 0.0,
+            "transition_accuracy": 0.36830357142857145,
+            "next_action_accuracy": 0.013392857142857142,
+            "contact_accuracy": 0.32142857142857145,
+            "object_micro_f1": 0.13704276146316333,
+            "held_out_episode_count": 14
+          },
+          "history": [],
+          "is_current": true,
+          "weights_repository": "none for this run: staged base nv-community/Cosmos3-Super weights were evaluated through vLLM; create a separate repo only after new adapter or fine-tuned weights exist"
+        },
         {
           "id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval",
           "title": "Qwen3-Omni LoRA",
@@ -508,11 +550,63 @@
               "global_step": 712
             }
           ],
+          "is_current": false,
+          "weights_repository": "historical diagnostic package; keep separate from the final 128-episode adapter repo"
+        },
+        {
+          "id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "title": "Qwen3-Omni LoRA",
+          "status": "verified",
+          "backbone": "qwen3_omni_lora",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "structured_episode_understanding_json_qa",
+          "source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora",
+          "eval_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 1.0,
+            "action_macro_f1": 0.0021983997167007384,
+            "subtask_accuracy": 0.002232142857142857,
+            "transition_accuracy": 0.9732142857142857,
+            "next_action_accuracy": 0.03125,
+            "contact_accuracy": 0.7209821428571429,
+            "object_micro_f1": 0.30688228657389993,
+            "held_out_episode_count": 14
+          },
+          "history": [
+            {
+              "epoch": 1,
+              "train_loss": 0.41282760031950355,
+              "val_loss": 0.03288277983665466,
+              "global_step": 356
+            },
+            {
+              "epoch": 2,
+              "train_loss": 0.027745448225544075,
+              "val_loss": 0.027823254466056824,
+              "global_step": 712
+            }
+          ],
           "is_current": true,
           "weights_repository": "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep"
         }
       ],
-      "interpretation": "This layer contains the held-out foundation-model packages. Qwen3-Omni packages evaluate structured JSON task prediction; Cosmos3-Nano currently evaluates a future-window world-model compatibility adapter, not a full diffusion-weight fine-tune."
+      "interpretation": "This layer contains the held-out foundation-model packages. Qwen3-Omni packages evaluate structured JSON task prediction; Cosmos3-Nano evaluates a future-window world-model compatibility adapter; Cosmos3-Super Reasoner evaluates staged base weights through vLLM on the JSON task. Neither Cosmos branch is a new fine-tuned weight release yet."
     }
   ],
   "model_groups": [
@@ -745,6 +839,58 @@
               "global_step": 712
             }
           ],
+          "is_current": false,
+          "weights_repository": "historical diagnostic package; keep separate from the final 128-episode adapter repo"
+        },
+        {
+          "id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "title": "Qwen3-Omni LoRA",
+          "status": "verified",
+          "backbone": "qwen3_omni_lora",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "structured_episode_understanding_json_qa",
+          "source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora",
+          "eval_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 1.0,
+            "action_macro_f1": 0.0021983997167007384,
+            "subtask_accuracy": 0.002232142857142857,
+            "transition_accuracy": 0.9732142857142857,
+            "next_action_accuracy": 0.03125,
+            "contact_accuracy": 0.7209821428571429,
+            "object_micro_f1": 0.30688228657389993,
+            "held_out_episode_count": 14
+          },
+          "history": [
+            {
+              "epoch": 1,
+              "train_loss": 0.41282760031950355,
+              "val_loss": 0.03288277983665466,
+              "global_step": 356
+            },
+            {
+              "epoch": 2,
+              "train_loss": 0.027745448225544075,
+              "val_loss": 0.027823254466056824,
+              "global_step": 712
+            }
+          ],
           "is_current": true,
           "weights_repository": "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep"
         }
@@ -815,16 +961,113 @@
         }
       ],
       "comparison_note": "The current 128-episode Cosmos result is a public-safe future-window compatibility adapter. It is not yet a full Cosmos diffusion/LoRA weight release."
+    },
+    {
+      "id": "cosmos3_super_reasoner",
+      "model_family": "Cosmos3-Super Reasoner",
+      "model_type": "base-weight vLLM Reasoner evaluation over nv-community/Cosmos3-Super",
+      "weight_repository": "none for this run; staged base weights only, no new fine-tuned weights",
+      "one_episode_runs": [
+        {
+          "id": "cosmos3_super_one_episode",
+          "title": "Cosmos3-Super One-Episode Fine-Tune",
+          "scope": "one public Xperience-10M sample episode",
+          "status": "not_run",
+          "source": null,
+          "weights": "none",
+          "interpretation": "No one-episode Cosmos3-Super adapter or fine-tuned weight run is published. The available Super result is the 128-episode held-out base-weight evaluation."
+        }
+      ],
+      "readiness_runs": [
+        {
+          "id": "xperience10m_cosmos3_super_training_readiness_20260607",
+          "title": "Cosmos3-Super Training Readiness Probe",
+          "scope": "selected 128-episode 96/16/16 JSON-task dataset and staged Cosmos3-Super runtime",
+          "status": "blocked_until_trainer_implemented",
+          "source": "results/omni_finetune/xperience10m_cosmos3_super_training_readiness_20260607/training_readiness.json",
+          "split": "train/val/test by selected episode/session",
+          "counts": {
+            "dataset_samples": 3808,
+            "split_counts": {
+              "test": {
+                "samples": 448,
+                "episodes": 14,
+                "actions": 189
+              },
+              "train": {
+                "samples": 2848,
+                "episodes": 89,
+                "actions": 885
+              },
+              "val": {
+                "samples": 512,
+                "episodes": 16,
+                "actions": 223
+              }
+            }
+          },
+          "primary_metrics": {
+            "diffusers_runtime_supported": true,
+            "chat_sft_supported": false,
+            "weights_updated": false
+          },
+          "weights": "none; readiness audit only, no adapter checkpoint",
+          "interpretation": "This probe confirms the staged Cosmos3-Super Diffusers/GPU runtime and the same JSON QA dataset are visible, but blocks true fine-tuning until a Cosmos-specific diffusion/action target packer and supervised loss are implemented."
+        }
+      ],
+      "multi_episode_128_runs": [
+        {
+          "id": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+          "title": "Cosmos3-Super Reasoner",
+          "status": "verified",
+          "backbone": "cosmos3_super_reasoner",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "zero_shot_structured_episode_understanding_json_qa_via_vllm_reasoner",
+          "source": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_cosmos3_super_reasoner_base_vllm_8gpu_20260607",
+          "eval_run_id": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 0.5111607142857143,
+            "action_macro_f1": 0.0008284021201089245,
+            "subtask_accuracy": 0.0,
+            "transition_accuracy": 0.36830357142857145,
+            "next_action_accuracy": 0.013392857142857142,
+            "contact_accuracy": 0.32142857142857145,
+            "object_micro_f1": 0.13704276146316333,
+            "held_out_episode_count": 14
+          },
+          "history": [],
+          "is_current": true,
+          "weights_repository": "none for this run: staged base nv-community/Cosmos3-Super weights were evaluated through vLLM; create a separate repo only after new adapter or fine-tuned weights exist"
+        }
+      ],
+      "comparison_note": "Cosmos3-Super is now represented by a verified 448-window held-out Reasoner evaluation on the same JSON task as Qwen3. It uses staged base weights through vLLM, so it is a model-branch diagnostic, not a weight release. The readiness probe records why true Cosmos3-Super fine-tuning is not launched yet."
     }
   ],
   "model_group_reading_notes": [
     "Use model_groups when comparing one-episode and 128-episode artifacts within the same model family.",
     "Task-head baselines have both a one-episode public-sample run and a 128-episode same-split metadata/text run.",
     "Qwen3-Omni has a one-episode sensor-adapter smoke test and separate 128-episode LoRA diagnostic packages; only the final 128-episode adapter belongs in the Qwen LoRA model repo.",
-    "Cosmos3-Nano currently has only a 128-episode future-window compatibility package; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist."
+    "Cosmos3-Nano has a 128-episode future-window compatibility package.",
+    "Cosmos3-Super has a 128-episode base-weight Reasoner evaluation on the JSON task plus a training-readiness probe; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist."
   ],
   "pending": [
     "Use the final Qwen3 full-eval package as the current Qwen result; older Qwen package rows remain historical diagnostics for comparison.",
-    "Promote Cosmos3 from compatibility adapter to full Cosmos3 fine-tuning only after a separate environment with matching Diffusers/Cosmos dependencies is prepared."
+    "Promote Cosmos3 from Nano compatibility and Super base-weight evaluation to true fine-tuning only after a dedicated Cosmos diffusion/action target packer and supervised loss produce new weights."
   ]
 }

data/project_packet.json

@@ -41,7 +41,7 @@
         "docs/data/scope_claims_audit.json",
         "docs/data/website_integrity.json"
       ],
-      "readout": "The project status table and roadmap give the compact current-state summary. Single-episode task engineering, metrics, visualizations, public website integrity, mirror parity, same-split 128-episode baselines, the final selected-episode Qwen3-Omni diagnostic result, and the Cosmos3-Nano compatibility package are implemented; stronger action/subtask and full Cosmos model quality remain follow-ups."
+      "readout": "The project status table and roadmap give the compact current-state summary. Single-episode task engineering, metrics, visualizations, public website integrity, mirror parity, same-split 128-episode baselines, the final selected-episode Qwen3-Omni diagnostic result, the Cosmos3-Nano compatibility package, and the Cosmos3-Super base-weight Reasoner evaluation are implemented; stronger action/subtask and real Cosmos fine-tuned model quality remain follow-ups."
     },
     {
       "step": 2,

data/project_status.json

@@ -31,6 +31,9 @@
     "qwen3_omni_lora_adapter_repo": "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep",
     "cosmos3_nano_future_window_compatibility_verified": true,
     "cosmos3_nano_future_window_test_predictions": 378,
+    "cosmos3_super_reasoner_verified": true,
+    "cosmos3_super_reasoner_test_predictions": 448,
+    "cosmos3_super_reasoner_json_validity_rate": 0.5111607142857143,
     "omni_model_comparison_available": true,
     "multi_episode_128_aligned_baselines": true,
     "multi_episode_128_baseline_window_counts": {
@@ -116,7 +119,7 @@
         "FOUNDATION_MODEL_PLAN.md",
         "docs/data/foundation_model_plan.json"
       ],
-      "readout": "Qwen3-Omni remains the first trainable held-out LoRA baseline; Cosmos 3 is now represented by a verified Cosmos3-Nano future-window compatibility package and remains the first world-model/action-generation branch; OpenVLA/openpi/GR00T are policy candidates after action targets are explicit."
+      "readout": "Qwen3-Omni remains the first trainable held-out LoRA baseline; Cosmos 3 is now represented by a verified Cosmos3-Nano future-window compatibility package plus a verified Cosmos3-Super base-weight Reasoner evaluation; OpenVLA/openpi/GR00T are policy candidates after action targets are explicit."
     },
     {
       "area": "Omni model extension contract",
@@ -204,20 +207,20 @@
         "results/omni_finetune/OMNI_MODEL_COMPARISON.md",
         "scripts/omni/build_omni_model_comparison.py"
       ],
-      "readout": "The public comparison now has two views: the three result layers and a model-family grouping. The model grouping pairs 1-episode and 128-episode entries for task-head baselines, separates Qwen3-Omni sensor-adapter smoke from 128-episode LoRA diagnostics, and marks Cosmos3-Nano as 128-episode compatibility-only until real Cosmos weights exist."
+      "readout": "The public comparison now has two views: the three result layers and a model-family grouping. The model grouping pairs 1-episode and 128-episode entries for task-head baselines, separates Qwen3-Omni sensor-adapter smoke from 128-episode LoRA diagnostics, and separates Cosmos3-Nano future-window compatibility from Cosmos3-Super base-weight Reasoner evaluation."
     },
     {
       "area": "Qwen3-Omni fine-tuning",
       "status": "final_verified_diagnostic_result_json_target_met",
       "evidence": [
         "docs/data/omni_finetune_verified_result.json",
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/",
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/",
         "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep",
         "scripts/omni/package_verified_omni_result.py",
         "scripts/omni/audit_verified_omni_package.py",
         "scripts/omni/analyze_qwen3_omni_errors.py"
       ],
-      "readout": "The selected 96/16/16 episode split produced a final public-safe held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs, validation/audit summaries, and a public LoRA adapter repo. JSON validity is 99.78%, meeting the 98% target; transition accuracy is 97.10%, contact accuracy is 71.88%, object micro-F1 is 30.16%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim."
+      "readout": "The selected 96/16/16 episode split now has a v3 strict-label public-safe held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs reused from the same LoRA adapter, validation/audit summaries, and a public LoRA adapter repo. JSON validity is 100.00%, meeting the 98% target; transition accuracy is 97.32%, contact accuracy is 72.10%, object micro-F1 is 30.69%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim."
     },
     {
       "area": "Cosmos3-Nano future-window branch",
@@ -230,6 +233,17 @@
       ],
       "readout": "The Cosmos3-Nano branch now has a public-safe verified future-window compatibility package with 3,213 future-window samples, 378 held-out test predictions, future retrieval MRR 0.0221, temporal consistency 0.0952, transition accuracy 0.9683, and contact accuracy 0.7434. It is a compatibility adapter result, not a full Cosmos diffusion-weight fine-tune."
     },
+    {
+      "area": "Cosmos3-Super Reasoner branch",
+      "status": "verified_base_weight_result",
+      "evidence": [
+        "configs/omni_backbones/cosmos3_super_reasoner.json",
+        "scripts/omni/eval_cosmos3_super_reasoner.py",
+        "scripts/omni/run_cosmos3_super_reasoner_eval.sh",
+        "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json"
+      ],
+      "readout": "Cosmos3-Super Reasoner now has a public-safe verified 448-window held-out evaluation on the same structured JSON task as Qwen3. It uses staged nv-community/Cosmos3-Super base weights through an 8-GPU vLLM server, not fine-tuned weights: JSON validity 0.5112, action macro-F1 0.0008, transition accuracy 0.3683, contact accuracy 0.3214, and object micro-F1 0.1370."
+    },
     {
       "area": "Raw Xperience-10M redistribution",
       "status": "not_included",
@@ -259,10 +273,10 @@
   ],
   "current_reading_notes": [
     "The final Qwen3-Omni diagnostic result is verified and meets the strict-JSON target, but action/subtask held-out quality is still weak.",
-    "Use docs/data/omni_model_comparison.json to compare both views: the single-episode/128-baseline/model-branch result layers and the model-family grouping for task heads, Qwen3-Omni LoRA, and Cosmos3-Nano.",
+    "Use docs/data/omni_model_comparison.json to compare both views: the single-episode/128-baseline/model-branch result layers and the model-family grouping for task heads, Qwen3-Omni LoRA, Cosmos3-Nano, and Cosmos3-Super.",
     "Use docs/data/omni_finetune_verified_result.json and the latest verified_public final Qwen package for current held-out results.",
     "The 128-episode aligned simple/NN baselines use metadata/text features from the derived Qwen JSONL export; they align the split and task ids but do not replace raw-modality baselines for trajectory, retrieval, reconstruction, or misalignment tasks.",
-    "The Cosmos3-Nano future-window branch is verified as a compatibility adapter result; one-episode Cosmos fine-tuning and full Cosmos diffusion-weight fine-tuning remain pending, so no Cosmos weight repo should be published yet.",
+    "The Cosmos3-Nano future-window branch is verified as a compatibility adapter result, and Cosmos3-Super Reasoner is verified as a base-weight evaluation; one-episode Cosmos fine-tuning and full Cosmos adapter/diffusion-weight fine-tuning remain pending, so no Cosmos weight repo should be published yet.",
     "The current reconstruction task reconstructs feature vectors, not pixel-depth, mesh, NeRF, or Gaussian reconstruction.",
     "Audio is one of the synchronized source modalities in the current task representation.",
     "The audio ablation report compares audio/no-audio variants across all 12 task contracts in results/audio_ablation/.",

data/publication_audit.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:41:02+00:00",
+  "generated_at_utc": "2026-06-07T15:49:07+00:00",
   "checks": [
     {
       "name": "required_publication_assets_present",
@@ -182,8 +182,8 @@
     "github_repo": {
       "root": "repo",
       "exists": true,
-      "file_count": 618,
-      "text_file_count": 523,
+      "file_count": 680,
+      "text_file_count": 577,
       "largest_file": {
         "path": "tmp/omni_128_dataset_fetch/dataset.jsonl",
         "bytes": 582271586
@@ -193,8 +193,8 @@
     "hf_space_bundle": {
       "root": "hf_publish/space",
       "exists": true,
-      "file_count": 519,
-      "text_file_count": 425,
+      "file_count": 582,
+      "text_file_count": 480,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -204,8 +204,8 @@
     "hf_artifact_bundle": {
       "root": "hf_publish/artifacts",
       "exists": true,
-      "file_count": 695,
-      "text_file_count": 577,
+      "file_count": 757,
+      "text_file_count": 631,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -215,8 +215,8 @@
     "hf_model_bundle": {
       "root": "hf_publish/model",
       "exists": true,
-      "file_count": 883,
-      "text_file_count": 730,
+      "file_count": 945,
+      "text_file_count": 784,
       "largest_file": {
         "path": "pytorch_model.bin",
         "bytes": 93495480

data/reproducibility_matrix.json

@@ -82,7 +82,7 @@
       "status": "verified_final_diagnostic_result_not_publicly_rerunnable_without_gated_data",
       "command": "scripts/omni/build_qwen3_omni_dataset.py and scripts/omni/train_qwen3_omni_lora.py on the selected gated episodes",
       "expected": "verified final diagnostic LoRA package with 3,808 exported windows, 2,848 train windows, and 448 held-out test predictions",
-      "boundary": "the public package records metrics and manifests, but rerunning requires gated Xperience-10M episode access and base-model weights; current JSON validity is 99.78%, meeting the 98% target, while action/subtask metrics remain weak"
+      "boundary": "the public package records metrics and manifests, but rerunning requires gated Xperience-10M episode access and base-model weights; current strict-label JSON validity is 100.00%, meeting the 98% target, while action/subtask metrics remain weak"
     }
   ]
 }

data/scope_claims_audit.json

@@ -1,15 +1,15 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T00:00:35+00:00",
+  "generated_at_utc": "2026-06-07T15:47:31+00:00",
   "summary": {
     "qwen3_omni_verified_diagnostic_pilot": true,
     "dataset_manifest_num_episodes": 119,
     "dataset_manifest_num_samples": 3808,
     "training_metadata_num_train_samples": 2848,
     "eval_num_samples": 448,
-    "eval_json_validity_rate": 0.9977678571428571,
+    "eval_json_validity_rate": 1.0,
     "quality_target_met": true,
-    "historical_identifier_count": 189,
+    "historical_identifier_count": 1545,
     "public_32_episode_status_file_count": 1,
     "failure_count": 0
   },
@@ -35,7 +35,7 @@
       "status": "pass",
       "detail": "episodes=119, samples=3808, split_counts={'train': 2848, 'val': 512, 'test': 448}",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/dataset/dataset_manifest.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json"
       ]
     },
     {
@@ -43,15 +43,15 @@
       "status": "pass",
       "detail": "train=2848, val=512, processes=8",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/training/training_metadata.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json"
       ]
     },
     {
       "name": "verified_package_eval_records_real_held_out_metrics",
       "status": "pass",
-      "detail": "samples=448, split=test, held_out=14, json_validity=0.9977678571428571",
+      "detail": "samples=448, split=test, held_out=14, json_validity=1.0",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/eval/metrics.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json"
       ]
     },
     {
@@ -59,7 +59,7 @@
       "status": "pass",
       "detail": "audit_status=pass, issues=0",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/package_audit.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json"
       ]
     },
     {
@@ -84,7 +84,7 @@
     {
       "name": "historical_32ep_identifiers_are_confined_to_readiness_artifacts",
       "status": "pass",
-      "detail": "historical identifiers found in result provenance files=189",
+      "detail": "historical identifiers found in result provenance files=1545",
       "evidence": [
         "results/omni_finetune/"
       ]
@@ -424,6 +424,6 @@
       "example": "{\"id\": \"xperience-10m-sample:qa:53\", \"episode_id\": \"xperience-10m-sample\", \"split\": \"train\", \"target\": \"episode_qa\", \"prompt_type\": \"json_episode_understanding\", \"center_window\": {\"start_frame\": 1060, \"end_frame\": 1079, \"num_frames\": 20}, \"media\": {\"video_path"
     }
   ],
-  "historical_identifier_total_count": 189,
+  "historical_identifier_total_count": 1545,
   "failures": []
 }

data/task_surface_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-06T23:27:06+00:00",
+  "generated_at_utc": "2026-06-07T15:47:30+00:00",
   "summary": {
     "task_count": 12,
     "expected_task_count": 12,
@@ -64,15 +64,15 @@
       "observed": "timeline_action"
     },
     {
-      "name": "timeline_action: public_field_plain_goal_is_human_readable",
+      "name": "timeline_action: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Look at one short multimodal window and name what action is happening now.",
+      "value": "20-frame multimodal window",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_card_blurb_is_human_readable",
+      "name": "timeline_action: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Recognize the current manipulation action from synchronized visual, motion, inertial, pose, and annotation context.",
+      "value": "Egocentric Action Recognition",
       "raw_hits": []
     },
     {
@@ -82,27 +82,27 @@
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_input_short_is_human_readable",
+      "name": "timeline_action: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "20-frame multimodal window",
+      "value": "Recognize the current manipulation action from synchronized visual, motion, inertial, pose, and annotation context.",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_output_short_is_human_readable",
+      "name": "timeline_action: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "current action class",
+      "value": "window features -> action label builder -> classifier",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_process_short_is_human_readable",
+      "name": "timeline_action: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "window features -> action label builder -> classifier",
+      "value": "current action class",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_research_name_is_human_readable",
+      "name": "timeline_action: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Egocentric Action Recognition",
+      "value": "Look at one short multimodal window and name what action is happening now.",
       "raw_hits": []
     },
     {
@@ -184,15 +184,15 @@
       "observed": "timeline_subtask"
     },
     {
-      "name": "timeline_subtask: public_field_plain_goal_is_human_readable",
+      "name": "timeline_subtask: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict the higher-level task stage for the current window.",
+      "value": "20-frame multimodal window",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_card_blurb_is_human_readable",
+      "name": "timeline_subtask: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Recognize the broader activity stage so fine actions become a readable procedure timeline.",
+      "value": "Temporal Subtask Recognition",
       "raw_hits": []
     },
     {
@@ -202,27 +202,27 @@
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_input_short_is_human_readable",
+      "name": "timeline_subtask: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "20-frame multimodal window",
+      "value": "Recognize the broader activity stage so fine actions become a readable procedure timeline.",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_output_short_is_human_readable",
+      "name": "timeline_subtask: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "current procedure step",
+      "value": "window features -> subtask label builder -> classifier",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_process_short_is_human_readable",
+      "name": "timeline_subtask: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "window features -> subtask label builder -> classifier",
+      "value": "current procedure step",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_research_name_is_human_readable",
+      "name": "timeline_subtask: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Subtask Recognition",
+      "value": "Predict the higher-level task stage for the current window.",
       "raw_hits": []
     },
     {
@@ -304,15 +304,15 @@
       "observed": "transition_detection"
     },
     {
-      "name": "transition_detection: public_field_plain_goal_is_human_readable",
+      "name": "transition_detection: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Detect whether the current window is near a boundary between actions.",
+      "value": "current window with boundary target",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_card_blurb_is_human_readable",
+      "name": "transition_detection: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Detect the local moment where the episode changes from one action segment to the next.",
+      "value": "Temporal Action Segmentation",
       "raw_hits": []
     },
     {
@@ -322,27 +322,27 @@
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_input_short_is_human_readable",
+      "name": "transition_detection: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current window with boundary target",
+      "value": "Detect the local moment where the episode changes from one action segment to the next.",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_output_short_is_human_readable",
+      "name": "transition_detection: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "boundary or steady",
+      "value": "action changes -> boundary labels -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_process_short_is_human_readable",
+      "name": "transition_detection: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "action changes -> boundary labels -> binary classifier",
+      "value": "boundary or steady",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_research_name_is_human_readable",
+      "name": "transition_detection: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Action Segmentation",
+      "value": "Detect whether the current window is near a boundary between actions.",
       "raw_hits": []
     },
     {
@@ -422,15 +422,15 @@
       "observed": "next_action"
     },
     {
-      "name": "next_action: public_field_plain_goal_is_human_readable",
+      "name": "next_action: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Use the current window to guess the action that will happen shortly after it.",
+      "value": "current window at time t",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_card_blurb_is_human_readable",
+      "name": "next_action: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Forecast the near-future action from the current observations only.",
+      "value": "Short-Horizon Intention Prediction",
       "raw_hits": []
     },
     {
@@ -440,27 +440,27 @@
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_input_short_is_human_readable",
+      "name": "next_action: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current window at time t",
+      "value": "Forecast the near-future action from the current observations only.",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_output_short_is_human_readable",
+      "name": "next_action: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "action at t+20 frames",
+      "value": "current features -> future label shift -> classifier",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_process_short_is_human_readable",
+      "name": "next_action: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "current features -> future label shift -> classifier",
+      "value": "action at t+20 frames",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_research_name_is_human_readable",
+      "name": "next_action: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Short-Horizon Intention Prediction",
+      "value": "Use the current window to guess the action that will happen shortly after it.",
       "raw_hits": []
     },
     {
@@ -540,15 +540,15 @@
       "observed": "hand_trajectory_forecast"
     },
     {
-      "name": "hand_trajectory_forecast: public_field_plain_goal_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict where the hands will move over the next few frames.",
+      "value": "current multimodal window",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_card_blurb_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict the future 3D left/right hand path from the current multimodal state.",
+      "value": "3D Hand Motion Forecasting",
       "raw_hits": []
     },
     {
@@ -558,27 +558,27 @@
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_input_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current multimodal window",
+      "value": "Predict the future 3D left/right hand path from the current multimodal state.",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_output_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "future hand-joint trajectory",
+      "value": "current features -> future mocap target -> regression head",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_process_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "current features -> future mocap target -> regression head",
+      "value": "future hand-joint trajectory",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_research_name_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "3D Hand Motion Forecasting",
+      "value": "Predict where the hands will move over the next few frames.",
       "raw_hits": []
     },
     {
@@ -658,15 +658,15 @@
       "observed": "contact_prediction"
     },
     {
-      "name": "contact_prediction: public_field_plain_goal_is_human_readable",
+      "name": "contact_prediction: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict whether the body or hand is in contact with something.",
+      "value": "non-contact, non-caption features",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_card_blurb_is_human_readable",
+      "name": "contact_prediction: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict whether body or hand contact with the scene is occurring without leaking contact labels.",
+      "value": "Human-Object Contact Prediction",
       "raw_hits": []
     },
     {
@@ -676,27 +676,27 @@
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_input_short_is_human_readable",
+      "name": "contact_prediction: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "non-contact, non-caption features",
+      "value": "Predict whether body or hand contact with the scene is occurring without leaking contact labels.",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_output_short_is_human_readable",
+      "name": "contact_prediction: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "contact or no contact",
+      "value": "feature filter -> contact target -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_process_short_is_human_readable",
+      "name": "contact_prediction: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "feature filter -> contact target -> binary classifier",
+      "value": "contact or no contact",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_research_name_is_human_readable",
+      "name": "contact_prediction: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Human-Object Contact Prediction",
+      "value": "Predict whether the body or hand is in contact with something.",
       "raw_hits": []
     },
     {
@@ -774,15 +774,15 @@
       "observed": "object_relevance"
     },
     {
-      "name": "object_relevance: public_field_plain_goal_is_human_readable",
+      "name": "object_relevance: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict which objects matter in the current window.",
+      "value": "non-caption multimodal features",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_card_blurb_is_human_readable",
+      "name": "object_relevance: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Infer which objects are relevant to the current manipulation window from non-caption features.",
+      "value": "Object-Centric Interaction Recognition",
       "raw_hits": []
     },
     {
@@ -792,27 +792,27 @@
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_input_short_is_human_readable",
+      "name": "object_relevance: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "non-caption multimodal features",
+      "value": "Infer which objects are relevant to the current manipulation window from non-caption features.",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_output_short_is_human_readable",
+      "name": "object_relevance: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "relevant object set",
+      "value": "object vocabulary -> multi-hot labels -> sigmoid heads",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_process_short_is_human_readable",
+      "name": "object_relevance: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "object vocabulary -> multi-hot labels -> sigmoid heads",
+      "value": "relevant object set",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_research_name_is_human_readable",
+      "name": "object_relevance: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Object-Centric Interaction Recognition",
+      "value": "Predict which objects matter in the current window.",
       "raw_hits": []
     },
     {
@@ -892,15 +892,15 @@
       "observed": "caption_grounding"
     },
     {
-      "name": "caption_grounding: public_field_plain_goal_is_human_readable",
+      "name": "caption_grounding: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Given a text-like query from annotation, find the matching time window.",
+      "value": "text-like query and candidate windows",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_card_blurb_is_human_readable",
+      "name": "caption_grounding: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Retrieve the matching time window for an annotation-derived text query.",
+      "value": "Language-to-Moment Grounding",
       "raw_hits": []
     },
     {
@@ -910,27 +910,27 @@
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_input_short_is_human_readable",
+      "name": "caption_grounding: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "text-like query and candidate windows",
+      "value": "Retrieve the matching time window for an annotation-derived text query.",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_output_short_is_human_readable",
+      "name": "caption_grounding: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "ranked matching moments",
+      "value": "query features -> candidate index -> cosine ranker",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_process_short_is_human_readable",
+      "name": "caption_grounding: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "query features -> candidate index -> cosine ranker",
+      "value": "ranked matching moments",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_research_name_is_human_readable",
+      "name": "caption_grounding: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Language-to-Moment Grounding",
+      "value": "Given a text-like query from annotation, find the matching time window.",
       "raw_hits": []
     },
     {
@@ -1008,15 +1008,15 @@
       "observed": "cross_modal_retrieval"
     },
     {
-      "name": "cross_modal_retrieval: public_field_plain_goal_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Use one group of modalities to retrieve the matching window from another group.",
+      "value": "motion/IMU/pose query; depth/video candidates",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_card_blurb_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Use motion, IMU, and camera-pose signals to retrieve the matching depth/video window.",
+      "value": "Multimodal Representation Retrieval",
       "raw_hits": []
     },
     {
@@ -1026,27 +1026,27 @@
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_input_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion/IMU/pose query; depth/video candidates",
+      "value": "Use motion, IMU, and camera-pose signals to retrieve the matching depth/video window.",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_output_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "ranked visual windows",
+      "value": "modality split -> projection -> nearest-neighbor ranker",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_process_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "modality split -> projection -> nearest-neighbor ranker",
+      "value": "ranked visual windows",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_research_name_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Multimodal Representation Retrieval",
+      "value": "Use one group of modalities to retrieve the matching window from another group.",
       "raw_hits": []
     },
     {
@@ -1126,15 +1126,15 @@
       "observed": "modality_reconstruction"
     },
     {
-      "name": "modality_reconstruction: public_field_plain_goal_is_human_readable",
+      "name": "modality_reconstruction: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict one modality feature block from other modality blocks.",
+      "value": "motion, IMU, and camera/pose features",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_card_blurb_is_human_readable",
+      "name": "modality_reconstruction: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict compressed depth/video feature vectors from motion, IMU, and camera-pose features.",
+      "value": "Modality Feature Reconstruction",
       "raw_hits": []
     },
     {
@@ -1144,27 +1144,27 @@
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_input_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion, IMU, and camera/pose features",
+      "value": "Predict compressed depth/video feature vectors from motion, IMU, and camera-pose features.",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_output_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "reconstructed depth/video vector",
+      "value": "source-target split -> scaler -> regression head",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_process_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "source-target split -> scaler -> regression head",
+      "value": "reconstructed depth/video vector",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_research_name_is_human_readable",
+      "name": "modality_reconstruction: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Modality Feature Reconstruction",
+      "value": "Predict one modality feature block from other modality blocks.",
       "raw_hits": []
     },
     {
@@ -1244,15 +1244,15 @@
       "observed": "temporal_order"
     },
     {
-      "name": "temporal_order: public_field_plain_goal_is_human_readable",
+      "name": "temporal_order: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Tell whether two nearby windows are in the correct time order.",
+      "value": "two adjacent windows plus difference vector",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_card_blurb_is_human_readable",
+      "name": "temporal_order: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Tell whether two neighboring windows are in chronological order or reversed.",
+      "value": "Temporal Order Verification",
       "raw_hits": []
     },
     {
@@ -1262,27 +1262,27 @@
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_input_short_is_human_readable",
+      "name": "temporal_order: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "two adjacent windows plus difference vector",
+      "value": "Tell whether two neighboring windows are in chronological order or reversed.",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_output_short_is_human_readable",
+      "name": "temporal_order: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "correct or reversed",
+      "value": "pair builder -> feature combiner -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_process_short_is_human_readable",
+      "name": "temporal_order: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "pair builder -> feature combiner -> binary classifier",
+      "value": "correct or reversed",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_research_name_is_human_readable",
+      "name": "temporal_order: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Order Verification",
+      "value": "Tell whether two nearby windows are in the correct time order.",
       "raw_hits": []
     },
     {
@@ -1360,15 +1360,15 @@
       "observed": "misalignment_detection"
     },
     {
-      "name": "misalignment_detection: public_field_plain_goal_is_human_readable",
+      "name": "misalignment_detection: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Detect when modalities that should match are shifted out of sync.",
+      "value": "motion-side and visual/depth-side feature groups",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_card_blurb_is_human_readable",
+      "name": "misalignment_detection: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Detect whether motion and visual/depth streams have been artificially shifted out of sync.",
+      "value": "Cross-Modal Misalignment Detection",
       "raw_hits": []
     },
     {
@@ -1378,27 +1378,27 @@
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_input_short_is_human_readable",
+      "name": "misalignment_detection: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion-side and visual/depth-side feature groups",
+      "value": "Detect whether motion and visual/depth streams have been artificially shifted out of sync.",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_output_short_is_human_readable",
+      "name": "misalignment_detection: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "aligned or shifted",
+      "value": "aligned/shifted pairs -> feature combiner -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_process_short_is_human_readable",
+      "name": "misalignment_detection: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "aligned/shifted pairs -> feature combiner -> binary classifier",
+      "value": "aligned or shifted",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_research_name_is_human_readable",
+      "name": "misalignment_detection: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Cross-Modal Misalignment Detection",
+      "value": "Detect when modalities that should match are shifted out of sync.",
       "raw_hits": []
     },
     {

data/website_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:40:59+00:00",
+  "generated_at_utc": "2026-06-07T15:47:32+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
@@ -75,7 +75,7 @@
       "status": "pass",
       "reason": "The project overview should appear before the deeper progress ledger.",
       "overview_index": 67412,
-      "evidence_index": 90476
+      "evidence_index": 90477
     },
     {
       "name": "project_status_links_json",
@@ -153,8 +153,8 @@
       "status": "pass",
       "reason": "The evaluation protocol should appear before the deeper evidence ledger.",
       "overview_index": 67412,
-      "protocol_index": 87159,
-      "evidence_index": 90476
+      "protocol_index": 87160,
+      "evidence_index": 90477
     },
     {
       "name": "evaluation_protocol_links_json",
@@ -252,7 +252,7 @@
     },
     {
       "path": "data/artifact_index.json",
-      "bytes": 60162,
+      "bytes": 67401,
       "top_level_type": "dict"
     },
     {
@@ -292,7 +292,7 @@
     },
     {
       "path": "data/mirror_parity.json",
-      "bytes": 257049,
+      "bytes": 410374,
       "top_level_type": "dict"
     },
     {
@@ -302,12 +302,12 @@
     },
     {
       "path": "data/omni_finetune_verified_result.json",
-      "bytes": 3483,
+      "bytes": 3628,
       "top_level_type": "dict"
     },
     {
       "path": "data/omni_model_comparison.json",
-      "bytes": 42015,
+      "bytes": 48296,
       "top_level_type": "dict"
     },
     {
@@ -327,7 +327,7 @@
     },
     {
       "path": "data/project_status.json",
-      "bytes": 16400,
+      "bytes": 16455,
       "top_level_type": "dict"
     },
     {
@@ -352,7 +352,7 @@
     },
     {
       "path": "data/reproducibility_matrix.json",
-      "bytes": 5280,
+      "bytes": 5294,
       "top_level_type": "dict"
     },
     {
@@ -382,7 +382,7 @@
     },
     {
       "path": "data/scope_claims_audit.json",
-      "bytes": 21234,
+      "bytes": 21251,
       "top_level_type": "dict"
     },
     {

docs/data/artifact_index.json

@@ -1,19 +1,19 @@
 {
   "title": "Ropedia Xperience-10M Task Suite Artifact Index",
-  "generated_at_utc": "2026-06-06T23:27:35+00:00",
+  "generated_at_utc": "2026-06-07T15:47:31+00:00",
   "status": "pass",
-  "artifact_count": 118,
+  "artifact_count": 129,
   "missing": [],
   "by_kind": {
     "project_path": 14,
     "scaleup_contract": 7,
-    "scaleup_status": 16,
+    "scaleup_status": 20,
     "publication_workflow": 5,
     "project_scope": 1,
     "source_alignment": 5,
     "evaluation_protocol": 3,
     "result_interpretation": 5,
-    "metrics_source": 14,
+    "metrics_source": 18,
     "website_data": 3,
     "visual_evidence": 7,
     "quality_gate": 12,
@@ -31,8 +31,8 @@
     "generated_figure_assets": 1,
     "citation": 1,
     "license": 1,
-    "verified_public_package": 4,
-    "publication_audit": 3
+    "verified_public_package": 6,
+    "publication_audit": 4
   },
   "artifacts": [
     {
@@ -65,8 +65,8 @@
       "surface": "repo_hf",
       "shows": "Gives a compact current-state table for first-pass readers.",
       "exists": true,
-      "bytes": 9845,
-      "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+      "bytes": 9926,
+      "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
     },
     {
       "id": "project_status_json",
@@ -76,8 +76,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable copy of the current project status for website and HF mirrors.",
       "exists": true,
-      "bytes": 15049,
-      "sha256": "23873ed59f3a38f46e45b15a5965afbb1365d49eb359bd5089a4ba6bda990d3c"
+      "bytes": 16455,
+      "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
     },
     {
       "id": "research_roadmap",
@@ -87,8 +87,8 @@
       "surface": "repo_hf",
       "shows": "Defines the path from public-sample task development to multi-episode held-out evaluation and larger omni-model extensions.",
       "exists": true,
-      "bytes": 12194,
-      "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+      "bytes": 12233,
+      "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
     },
     {
       "id": "research_roadmap_json",
@@ -142,8 +142,8 @@
       "surface": "repo_hf",
       "shows": "Stores the implemented Qwen3-Omni LoRA contract and planned Cosmos-style world-model and VLA/policy branch contracts.",
       "exists": true,
-      "file_count": 3,
-      "bytes": 9203
+      "file_count": 4,
+      "bytes": 12613
     },
     {
       "id": "omni_backbone_registry_validator",
@@ -197,8 +197,8 @@
       "surface": "repo_hf",
       "shows": "Computes public-safe held-out error-analysis tables by episode, action family, train-seen status, required-modality state, and object category.",
       "exists": true,
-      "bytes": 15676,
-      "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+      "bytes": 16957,
+      "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
     },
     {
       "id": "multi_episode_128_baseline_script",
@@ -219,8 +219,8 @@
       "surface": "repo_hf",
       "shows": "Builds the upload-ready Hugging Face adapter folder from a verified Qwen3 LoRA result summary and adapter directory.",
       "exists": true,
-      "bytes": 9843,
-      "sha256": "636132a7d299db4d874ec797e34acd7e37eea69994c2d39afaafaec6587169a0"
+      "bytes": 10710,
+      "sha256": "c5f33f7030e6861ccab465b71d4c1b4209b8fbbcebdca882aa5c7e8d70e76b37"
     },
     {
       "id": "additional_development_directions",
@@ -274,8 +274,8 @@
       "surface": "website_hf",
       "shows": "Gives a short project path with scope status and public surfaces.",
       "exists": true,
-      "bytes": 7943,
-      "sha256": "ffd5da5fd2c2dc82fa1beb74335a51a33317923b3e7ee4864e2b5031082b0a42"
+      "bytes": 8005,
+      "sha256": "2258fecb80850c745e60cb28733869c49a5182879d9d0461b666a5575e3c1610"
     },
     {
       "id": "artifact_guide",
@@ -285,8 +285,8 @@
       "surface": "repo_hf",
       "shows": "Gives the human-readable map from project scope to data, tasks, platform mirrors, and scale-up status.",
       "exists": true,
-      "bytes": 17508,
-      "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+      "bytes": 17930,
+      "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
     },
     {
       "id": "official_dataset_card_alignment",
@@ -686,7 +686,7 @@
       "volatile": true,
       "shows": "Records the last live GitHub/HF URL verification after upload.",
       "exists": true,
-      "bytes": 68749,
+      "bytes": 74880,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -697,8 +697,8 @@
       "surface": "repo",
       "shows": "Fetches the published GitHub/HF URLs and compares live hashes and public-card markers against the release assets.",
       "exists": true,
-      "bytes": 36847,
-      "sha256": "07fd059a9ff8c13b073f349c79f1f7d3abe839559cf0809e291f6ea9bbad21e8"
+      "bytes": 36201,
+      "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
     },
     {
       "id": "reproducibility_contract",
@@ -719,8 +719,8 @@
       "surface": "website_hf",
       "shows": "Machine-readable reproduction steps with expected artifacts and public boundaries.",
       "exists": true,
-      "bytes": 5280,
-      "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+      "bytes": 5294,
+      "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
     },
     {
       "id": "artifact_index_builder",
@@ -754,7 +754,7 @@
       "volatile": true,
       "shows": "Separates setup paths from completed held-out-episode results.",
       "exists": true,
-      "bytes": 21234,
+      "bytes": 21251,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -766,7 +766,7 @@
       "volatile": true,
       "shows": "Confirms prepared GitHub/HF Space/artifact/model mirrors share the same critical data, figure, website HTML, and validator files.",
       "exists": true,
-      "bytes": 235815,
+      "bytes": 410374,
       "hash_policy": "existence_and_size_only"
     },
     {
@@ -965,8 +965,8 @@
       "surface": "repo_hf",
       "shows": "Documents the final 128-episode LoRA adapter upload path, target model repo, package builder, and forbidden files.",
       "exists": true,
-      "bytes": 1875,
-      "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+      "bytes": 1886,
+      "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
     },
     {
       "id": "multi_episode_access_status",
@@ -1031,8 +1031,8 @@
       "surface": "repo_hf",
       "shows": "Reader-facing comparison of the single-episode task suite, 128-episode aligned baselines, Qwen3-Omni packages, and Cosmos3 future-window branch.",
       "exists": true,
-      "bytes": 3110,
-      "sha256": "11c22b7ac1e16fd8db86eb7c6fc33cf28fee97a38098f1606a35daee113dc72b"
+      "bytes": 9231,
+      "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
     },
     {
       "id": "omni_model_comparison_json",
@@ -1042,8 +1042,8 @@
       "surface": "repo_hf",
       "shows": "Machine-readable comparison of the current result versions, per-task aligned baselines, verified Qwen3 packages, and Cosmos3 package.",
       "exists": true,
-      "bytes": 21433,
-      "sha256": "b539a489a8974ecec90dda312471be54f466b81bef9d1ebc99d08155f8c21c94"
+      "bytes": 48296,
+      "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
     },
     {
       "id": "cosmos3_nano_verified_summary",
@@ -1144,6 +1144,61 @@
       "bytes": 1099,
       "sha256": "f11ccb167908d4f5bfb49c0be0b4bc6c9254901462aa52ae98a2a98e8af16558"
     },
+    {
+      "id": "verified_public_package_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified public package: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "kind": "verified_public_package",
+      "surface": "repo_hf",
+      "shows": "Public-safe verified package for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607 (cosmos3_super_reasoner, status=verified).",
+      "exists": true,
+      "file_count": 21,
+      "bytes": 5151738
+    },
+    {
+      "id": "verified_public_summary_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified summary: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Machine-readable verified summary for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 5693,
+      "sha256": "6e263015a73dbcd8d9dd2fa036348d1a20c3e4aecfd1f6bf43d1b9f923320b91"
+    },
+    {
+      "id": "verified_public_public_result_summary_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified public result summary: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/PUBLIC_RESULT_SUMMARY.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Public result summary for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 979,
+      "sha256": "1c17ac8b08e522abaf606a32982086295c577c81f94a3495ea2b9b6a50f83d74"
+    },
+    {
+      "id": "verified_public_run_report_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified run report: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/eval/RUN_REPORT.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Run report for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 549,
+      "sha256": "9a3edf661c75956335137f98fa4cd40071bcc0a66e7553bb614375f2e2e2ac22"
+    },
+    {
+      "id": "verified_public_metrics_JSON_xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+      "title": "Verified metrics JSON: Cosmos3-Super Reasoner",
+      "path": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/eval/metrics.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Metrics json for xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607.",
+      "exists": true,
+      "bytes": 48156,
+      "sha256": "138b9de0ba2cda7b273c47a59656f5abc85c435a64108d9932bc3611143e5d8e"
+    },
     {
       "id": "verified_public_package_xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval",
       "title": "Verified public package: Qwen3-Omni LoRA",
@@ -1284,8 +1339,8 @@
       "surface": "repo_hf",
       "shows": "Public-safe verified package for xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full (qwen3_omni_lora, status=verified).",
       "exists": true,
-      "file_count": 16,
-      "bytes": 4898687
+      "file_count": 23,
+      "bytes": 4923819
     },
     {
       "id": "verified_public_summary_xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full",
@@ -1341,6 +1396,72 @@
       "exists": true,
       "bytes": 623,
       "sha256": "d7264cfb34e48b5c41c89444ea9cd1314b8f4d0bcc0224debbbe5ea512450197"
+    },
+    {
+      "id": "verified_public_package_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified public package: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "kind": "verified_public_package",
+      "surface": "repo_hf",
+      "shows": "Public-safe verified package for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full (qwen3_omni_lora, status=verified).",
+      "exists": true,
+      "file_count": 23,
+      "bytes": 4879339
+    },
+    {
+      "id": "verified_public_summary_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified summary: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Machine-readable verified summary for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 6208,
+      "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+    },
+    {
+      "id": "verified_public_public_result_summary_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified public result summary: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Public result summary for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 1002,
+      "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+    },
+    {
+      "id": "verified_public_run_report_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified run report: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+      "kind": "scaleup_status",
+      "surface": "repo_hf",
+      "shows": "Run report for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 580,
+      "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+    },
+    {
+      "id": "verified_public_metrics_JSON_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified metrics JSON: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+      "kind": "metrics_source",
+      "surface": "repo_hf",
+      "shows": "Metrics json for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 47129,
+      "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+    },
+    {
+      "id": "verified_public_package_audit_xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+      "title": "Verified package audit: Qwen3-Omni LoRA",
+      "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+      "kind": "publication_audit",
+      "surface": "repo_hf",
+      "shows": "Package audit for xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full.",
+      "exists": true,
+      "bytes": 634,
+      "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
     }
   ]
 }

docs/data/mirror_parity.json

@@ -1,9 +1,9 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:41:47+00:00",
+  "generated_at_utc": "2026-06-07T15:49:31+00:00",
   "hf_root": "hf_publish",
   "summary": {
-    "group_count": 197,
+    "group_count": 234,
     "failure_count": 0,
     "failures_by_surface": {}
   },
@@ -102,27 +102,27 @@
       "local": {
         "path": "repo:docs/data/artifact_index.json",
         "exists": true,
-        "bytes": 60162,
-        "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+        "bytes": 67401,
+        "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         },
         "hf_model": {
           "path": "hf_model:metrics/artifact_index.json",
           "exists": true,
-          "bytes": 60162,
-          "sha256": "304aa2ce52c9f69cfbd9b240e252ad1646556a02e9e3175b03cb399be55831be"
+          "bytes": 67401,
+          "sha256": "4d5ae505ddbe2c29a17b680cbee2232bc80ae9e5825cebe9755ec97909df6602"
         }
       },
       "failures": []
@@ -289,26 +289,26 @@
         "path": "repo:docs/data/live_publication_status.json",
         "exists": true,
         "bytes": 74880,
-        "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+        "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         },
         "hf_model": {
           "path": "hf_model:metrics/live_publication_status.json",
           "exists": true,
           "bytes": 74880,
-          "sha256": "56dc103e2cf8a7636f06037b438ea45a4a259e960c77f7b5421b14b4c2dcab8e"
+          "sha256": "c08305aae29ffb62d8f4b8d7c0f889199718cfcbe824352b35b302428edacca1"
         }
       },
       "failures": []
@@ -350,27 +350,27 @@
       "local": {
         "path": "repo:docs/data/omni_finetune_verified_result.json",
         "exists": true,
-        "bytes": 3483,
-        "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+        "bytes": 3628,
+        "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         },
         "hf_model": {
           "path": "hf_model:metrics/omni_finetune_verified_result.json",
           "exists": true,
-          "bytes": 3483,
-          "sha256": "890cbbc334a39230d97ce7bf83156e860e09c4451d97e0fde628623464d92fdf"
+          "bytes": 3628,
+          "sha256": "ce28a11876aa33feb1f7b28c977c1d3e708b7d5d8b24b062684d472ba671d004"
         }
       },
       "failures": []
@@ -381,27 +381,27 @@
       "local": {
         "path": "repo:docs/data/omni_model_comparison.json",
         "exists": true,
-        "bytes": 42015,
-        "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+        "bytes": 48296,
+        "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         },
         "hf_model": {
           "path": "hf_model:metrics/omni_model_comparison.json",
           "exists": true,
-          "bytes": 42015,
-          "sha256": "f74673dcc879ba73f7bc3210d8716ae5ad5d7de70ee76d59f56bb24443669c12"
+          "bytes": 48296,
+          "sha256": "1c968bd58842af9a4e6159c1a8bd171aec08757bb77fce9f04c55030be08357f"
         }
       },
       "failures": []
@@ -505,27 +505,27 @@
       "local": {
         "path": "repo:docs/data/project_status.json",
         "exists": true,
-        "bytes": 16400,
-        "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+        "bytes": 16455,
+        "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         },
         "hf_model": {
           "path": "hf_model:metrics/project_status.json",
           "exists": true,
-          "bytes": 16400,
-          "sha256": "0ef6b75a18793906e3f0c45e3b37f4023c6a3af9fcf7367bfa43924f161b6eff"
+          "bytes": 16455,
+          "sha256": "3590ee1e09ecf819080a7714ea9629db305e1fd68c99a65f62bb65061c0d766c"
         }
       },
       "failures": []
@@ -537,26 +537,26 @@
         "path": "repo:docs/data/publication_audit.json",
         "exists": true,
         "bytes": 7212,
-        "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+        "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         },
         "hf_model": {
           "path": "hf_model:metrics/publication_audit.json",
           "exists": true,
           "bytes": 7212,
-          "sha256": "bbb46638856b3c9c5cbf752f0f9a5cbe44dfbf3f4d29633c945daafa62eb0d36"
+          "sha256": "cd273d9c34f2e6c987b9299947354e363e195bd3e26c481ba2affe71f823212b"
         }
       },
       "failures": []
@@ -660,27 +660,27 @@
       "local": {
         "path": "repo:docs/data/reproducibility_matrix.json",
         "exists": true,
-        "bytes": 5280,
-        "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+        "bytes": 5294,
+        "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         },
         "hf_model": {
           "path": "hf_model:metrics/reproducibility_matrix.json",
           "exists": true,
-          "bytes": 5280,
-          "sha256": "bfb34f14206943da909aee36465e8211c592615fca15a284e2fa8ef9ea1d438b"
+          "bytes": 5294,
+          "sha256": "b91da81ad855de974037ba24bd8b1d132be840b8b50004d9c137f4b5064ad18d"
         }
       },
       "failures": []
@@ -846,27 +846,27 @@
       "local": {
         "path": "repo:docs/data/scope_claims_audit.json",
         "exists": true,
-        "bytes": 21234,
-        "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+        "bytes": 21251,
+        "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         },
         "hf_model": {
           "path": "hf_model:metrics/scope_claims_audit.json",
           "exists": true,
-          "bytes": 21234,
-          "sha256": "49329072596d48d5717fbb1bf90b65e7c62e43186af672430536655540f32842"
+          "bytes": 21251,
+          "sha256": "ead427da73d65d0f43bf6740483a1ffc8fb39f0e1022e4595938cf399ce605bc"
         }
       },
       "failures": []
@@ -971,26 +971,26 @@
         "path": "repo:docs/data/task_surface_integrity.json",
         "exists": true,
         "bytes": 45779,
-        "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+        "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         },
         "hf_model": {
           "path": "hf_model:metrics/task_surface_integrity.json",
           "exists": true,
           "bytes": 45779,
-          "sha256": "b2f4fdb1384e610a8961d706db33b1ef1b25e9378d283c34d1a66e4ab0a95bab"
+          "sha256": "ece1f964ae3face2151489cc9a24b43e446968af07e6e87d5f531733c27dcc35"
         }
       },
       "failures": []
@@ -1033,26 +1033,26 @@
         "path": "repo:docs/data/website_integrity.json",
         "exists": true,
         "bytes": 15375,
-        "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+        "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:data/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:docs/data/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         },
         "hf_model": {
           "path": "hf_model:metrics/website_integrity.json",
           "exists": true,
           "bytes": 15375,
-          "sha256": "ee640e4a0b2271ed97b8c57ce6b23542b474c46d9baa77fdfb003c2f372b4975"
+          "sha256": "449b5525a0fc9ba200e59c6248e5ce963381938ab2c2027e1933db9483622037"
         }
       },
       "failures": []
@@ -1760,21 +1760,21 @@
       "local": {
         "path": "repo:scripts/omni/analyze_qwen3_omni_errors.py",
         "exists": true,
-        "bytes": 15676,
-        "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+        "bytes": 16957,
+        "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/analyze_qwen3_omni_errors.py",
           "exists": true,
-          "bytes": 15676,
-          "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+          "bytes": 16957,
+          "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/analyze_qwen3_omni_errors.py",
           "exists": true,
-          "bytes": 15676,
-          "sha256": "d4c7e46d9fbd5f9d84bc32374f457fd8c9d68c8faa39c77bc45770eb95d80337"
+          "bytes": 16957,
+          "sha256": "1d5fb2e0df0b89b1d1236ffc0eb30a2033fa1c3bc00abfcdf7898108ab1e3382"
         }
       },
       "failures": []
@@ -1785,21 +1785,21 @@
       "local": {
         "path": "repo:scripts/omni/build_omni_model_comparison.py",
         "exists": true,
-        "bytes": 27997,
-        "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+        "bytes": 30236,
+        "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/omni/build_omni_model_comparison.py",
           "exists": true,
-          "bytes": 27997,
-          "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+          "bytes": 30236,
+          "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
         },
         "hf_model": {
           "path": "hf_model:scripts/omni/build_omni_model_comparison.py",
           "exists": true,
-          "bytes": 27997,
-          "sha256": "6f59fa792f26a4834cf168af92f50bb5a1f5e4de7449360dae87afc51cf5e0de"
+          "bytes": 30236,
+          "sha256": "207b0bbfbea1cd3d7e6e77e7eafcf231b71c9f6483ffc36889234c7bafbcb1df"
         }
       },
       "failures": []
@@ -2160,21 +2160,21 @@
       "local": {
         "path": "repo:scripts/verify_live_publication.py",
         "exists": true,
-        "bytes": 36237,
-        "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+        "bytes": 36201,
+        "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
       },
       "mirrors": {
         "hf_artifacts": {
           "path": "hf_artifacts:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 36237,
-          "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+          "bytes": 36201,
+          "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
         },
         "hf_model": {
           "path": "hf_model:scripts/verify_live_publication.py",
           "exists": true,
-          "bytes": 36237,
-          "sha256": "ca3c79330d9c8f476aea564ac3d6c966562857fc128503802d2e6342e5d9e021"
+          "bytes": 36201,
+          "sha256": "76f03885867a8ed7095958a6948cbce81b4958fb74a09df24c24ad7eb5b0d944"
         }
       },
       "failures": []
@@ -2410,21 +2410,21 @@
       "local": {
         "path": "repo:docs/index.html",
         "exists": true,
-        "bytes": 180715,
-        "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+        "bytes": 180727,
+        "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:index.html",
           "exists": true,
-          "bytes": 180715,
-          "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+          "bytes": 180727,
+          "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
         },
         "hf_artifacts_docs": {
           "path": "hf_artifacts:docs/index.html",
           "exists": true,
-          "bytes": 180715,
-          "sha256": "c1a311421a1567227e36c53a0ec0a4e993b44bff2c7ff3e43790d5fca959f7d5"
+          "bytes": 180727,
+          "sha256": "a88769e505d5af34674278f282ed1f482cc91dc711ddc0ed894a3fca5d08ff67"
         }
       },
       "failures": []
@@ -2665,27 +2665,27 @@
       "local": {
         "path": "repo:results/omni_finetune/HF_UPLOAD.md",
         "exists": true,
-        "bytes": 1875,
-        "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+        "bytes": 1886,
+        "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/HF_UPLOAD.md",
           "exists": true,
-          "bytes": 1875,
-          "sha256": "7a822452347e8c4241a5160d67a9782f17f3d3cb9bd2960b00bac0ca1bf2392f"
+          "bytes": 1886,
+          "sha256": "a8fa5a326e2dd91130c2e86b988e7afcfa4e043e48614191f0fc589158089c03"
         }
       },
       "failures": []
@@ -2696,27 +2696,27 @@
       "local": {
         "path": "repo:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
         "exists": true,
-        "bytes": 8275,
-        "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+        "bytes": 9231,
+        "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         },
         "hf_model": {
           "path": "hf_model:results/omni_finetune/OMNI_MODEL_COMPARISON.md",
           "exists": true,
-          "bytes": 8275,
-          "sha256": "b30089afb6e54a32b64b79d48fdd995c79123da6dda11b6dca8eeb675bd25edb"
+          "bytes": 9231,
+          "sha256": "c38d12e138193f7200800d4dd8c149497de2c5f5895299e22fe81285b69fc62d"
         }
       },
       "failures": []
@@ -3279,6 +3279,223 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3361,
+        "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3361,
+          "sha256": "d6cc31528f7af5711f13cdc74c4b403874cf56387bb58f69de0862994ac52c4d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1225,
+        "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1225,
+          "sha256": "9decba3f77266ca5b697eac61978bf71539cc87c4b897bbe96466a67adc3372b"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 1994,
+        "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1994,
+          "sha256": "5a5d1808b9495044b7173bb33bb1344415e97d6140ed6998a8514e3c48eda357"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 21118,
+        "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 21118,
+          "sha256": "3209343a10c73ec6dcdbd0781cf920ba371e3a29029e9feb8c5c8ecc46c65bf2"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 393,
+        "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 393,
+          "sha256": "2a08157d30f0d15e18e5af6b8348931dfad7a956b889b00bcbe78f15c8ee5332"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1618,
+        "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1618,
+          "sha256": "0aadc6fd85bdd20891859e3754a00be6dcf354031ece525a8017711030a7b49d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 522,
+        "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 522,
+          "sha256": "e15bf2696ca261367e42d5606649e4437ed278e33551a69329bd420fe507affb"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/dataset/dataset_manifest.json",
       "status": "pass",
@@ -4860,6 +5077,223 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3348,
+        "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3348,
+          "sha256": "fd98bef2ed55443a9f06442ed5e98cb39584f9793c50d466ca2dcb22a1340aeb"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1223,
+        "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1223,
+          "sha256": "c6c9e6e35dddd49a2d84512c19f5df0b3ac7ec344757b55be5d727bf0765cd7c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 2044,
+        "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 2044,
+          "sha256": "41ebbd45f8e338ea4b9a9da6502304d8bef9b4b91c9ec7953f698a5ee452b8f9"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 15918,
+        "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15918,
+          "sha256": "60ce5c6b039465a7978115a0ba012d86688fd4db2b27544cbf639a30eebf6723"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 395,
+        "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 395,
+          "sha256": "02fdcbbe88fb9c6717751226097185800811bf79dc9b95533b7df93b51b1335a"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1663,
+        "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1663,
+          "sha256": "d99b2156d35f7121b4c810b0e41eeac02a24f584d33268c6c096937e560df393"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 541,
+        "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 541,
+          "sha256": "1203749da28a8e7678916b75c8d43d7dc9b5f83a29110256bedbf5c688d5c72c"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/dataset/dataset_manifest.json",
       "status": "pass",
@@ -5325,6 +5759,719 @@
       },
       "failures": []
     },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+        "exists": true,
+        "bytes": 1002,
+        "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md",
+          "exists": true,
+          "bytes": 1002,
+          "sha256": "c3a07b041ef66250cb805625473f919cb01cc427f7ca79d12fd5142094f4a80c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+        "exists": true,
+        "bytes": 3352,
+        "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md",
+          "exists": true,
+          "bytes": 3352,
+          "sha256": "cea36dd7e5f9840b1d801263699b5ede7f4e4584991edc30ea0046a87b79da69"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+        "exists": true,
+        "bytes": 1260,
+        "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv",
+          "exists": true,
+          "bytes": 1260,
+          "sha256": "1e910d65a0e61c9ad0b905d233067f698ca9c53a0c00238ede242de4c2324ddc"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+        "exists": true,
+        "bytes": 1967,
+        "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv",
+          "exists": true,
+          "bytes": 1967,
+          "sha256": "ab4dc015f016d11d1fab6f84f0a2a964694795396d3792e7c111372631a666f8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+        "exists": true,
+        "bytes": 15657,
+        "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json",
+          "exists": true,
+          "bytes": 15657,
+          "sha256": "e64e72267bcbcfac0f835e6c7ef1940b656a880226c886315a18c173a317868b"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+        "exists": true,
+        "bytes": 366,
+        "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv",
+          "exists": true,
+          "bytes": 366,
+          "sha256": "a5f3d33e822b37bf605958d6891e4d3c4437d89afce611bea7c4cddcd55d477c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+        "exists": true,
+        "bytes": 1655,
+        "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/object_category_error_analysis.csv",
+          "exists": true,
+          "bytes": 1655,
+          "sha256": "e23f2b6b457ba72e1766575524b2fa7eecc2b9b8163e41bf60280bdac3dc7280"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+        "exists": true,
+        "bytes": 521,
+        "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/train_seen_error_analysis.csv",
+          "exists": true,
+          "bytes": 521,
+          "sha256": "befe2421974b023bda56b18501efa4f8ea3bf19a7995c7453d732dbbeb4bd3f5"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+        "exists": true,
+        "bytes": 252195,
+        "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json",
+          "exists": true,
+          "bytes": 252195,
+          "sha256": "01d836079c7c9dd652a73cdde7bf50f9ca7b51b347a35c0a2e5239ea240e21d8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+        "exists": true,
+        "bytes": 753255,
+        "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+          "exists": true,
+          "bytes": 753255,
+          "sha256": "d47f7acf526e2328c7203f6d317c9b188abf03f69c3108fc47e5e7d9a543f638"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+        "exists": true,
+        "bytes": 580,
+        "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/RUN_REPORT.md",
+          "exists": true,
+          "bytes": 580,
+          "sha256": "84ad4fa629d0f0bc6ee7da284a793e2d78f0e847c093fdbfcd87cd5df014c7e8"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+        "exists": true,
+        "bytes": 2948548,
+        "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/confusion_matrix.csv",
+          "exists": true,
+          "bytes": 2948548,
+          "sha256": "6554ff4f3ba46a7650ee5cebbcf412709b01a9ee01cf2a0f2d01259cd381d3fb"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+        "exists": true,
+        "bytes": 47129,
+        "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json",
+          "exists": true,
+          "bytes": 47129,
+          "sha256": "567a6db75813d638dfcdecc18c661764ddf06d8a0ec01b23af6a5113c7f06026"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+        "exists": true,
+        "bytes": 46482,
+        "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/per_class_metrics.csv",
+          "exists": true,
+          "bytes": 46482,
+          "sha256": "9dc4060b6ae7355b0b403f54aa67d5ddc108cdc8b584e56691c69e9323dc7cb0"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+        "exists": true,
+        "bytes": 222559,
+        "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.csv",
+          "exists": true,
+          "bytes": 222559,
+          "sha256": "fa3984cfe9f1895aed9b1d2bc6f9e49fae70c8c13bbca2439323b4bf39e244ac"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+        "exists": true,
+        "bytes": 550610,
+        "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/predictions.jsonl",
+          "exists": true,
+          "bytes": 550610,
+          "sha256": "839b2319ec7aced82d5d1552668e9e45744dce79f0aa40ab05e4ce03b9a9fa2d"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+        "exists": true,
+        "bytes": 634,
+        "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json",
+          "exists": true,
+          "bytes": 634,
+          "sha256": "c5bfa7b231dcf2f9a2d1a50ce5324cdf4bb8afd60294fd5d40b7ef578983b1b2"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+        "exists": true,
+        "bytes": 508,
+        "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/adapter_shape_check.json",
+          "exists": true,
+          "bytes": 508,
+          "sha256": "b37cfc87f20db697e48309d3d032b71cfda0a43490b9c5b1ab5535f46580c70c"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+        "exists": true,
+        "bytes": 16360,
+        "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/progress.jsonl",
+          "exists": true,
+          "bytes": 16360,
+          "sha256": "600af5926eb2d61bac79272f267167dbec2d44be409b0ef9059077cc7bb53b35"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+        "exists": true,
+        "bytes": 2429,
+        "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json",
+          "exists": true,
+          "bytes": 2429,
+          "sha256": "a772b61bfa934d8cd7b092d85ba9c12f65f30e8e1f0d4c270c03af1070ee2669"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+        "exists": true,
+        "bytes": 3354,
+        "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/eval.json",
+          "exists": true,
+          "bytes": 3354,
+          "sha256": "c4a5ab631f3c12b14c9c6de50220fe7ad00285f2d027357584fe86d266747f18"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+        "exists": true,
+        "bytes": 2708,
+        "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/validation/training.json",
+          "exists": true,
+          "bytes": 2708,
+          "sha256": "cbc71bb531b8b7ef8a1a9c6f21ebce2fd5d8a2ac4cd912c4fc9822f01e880607"
+        }
+      },
+      "failures": []
+    },
+    {
+      "name": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+      "status": "pass",
+      "local": {
+        "path": "repo:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+        "exists": true,
+        "bytes": 6208,
+        "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+      },
+      "mirrors": {
+        "hf_space": {
+          "path": "hf_space:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        },
+        "hf_artifacts": {
+          "path": "hf_artifacts:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        },
+        "hf_model": {
+          "path": "hf_model:results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "exists": true,
+          "bytes": 6208,
+          "sha256": "4a2fa145882d06008ebde7734dce72e81e1a111e8a24cc2877694249847d70cc"
+        }
+      },
+      "failures": []
+    },
     {
       "name": "results/single_episode_diagnostics/README.md",
       "status": "pass",
@@ -5610,27 +6757,27 @@
       "local": {
         "path": "repo:ARTIFACT_GUIDE.md",
         "exists": true,
-        "bytes": 17508,
-        "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+        "bytes": 17930,
+        "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         },
         "hf_model": {
           "path": "hf_model:ARTIFACT_GUIDE.md",
           "exists": true,
-          "bytes": 17508,
-          "sha256": "fbbd9f460610464efb27c371a17cf23c3fa409d853f8148368f48707192427d7"
+          "bytes": 17930,
+          "sha256": "6a2843318d83b8dac1fc56a53e4dbf202f8757d4e7d8b3bf67b574be570fa585"
         }
       },
       "failures": []
@@ -5889,27 +7036,27 @@
       "local": {
         "path": "repo:RESEARCH_ROADMAP.md",
         "exists": true,
-        "bytes": 12194,
-        "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+        "bytes": 12233,
+        "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         },
         "hf_model": {
           "path": "hf_model:RESEARCH_ROADMAP.md",
           "exists": true,
-          "bytes": 12194,
-          "sha256": "8773f240e362198b3a669d1ac848d6f1629df3a33e41bd76fba157cbf566479c"
+          "bytes": 12233,
+          "sha256": "020512aa647cef7d63eccf7bb8dd6cb86f0e5c457f3c0e3d5ef293e7b35a58bf"
         }
       },
       "failures": []
@@ -5920,27 +7067,27 @@
       "local": {
         "path": "repo:PROJECT_STATUS.md",
         "exists": true,
-        "bytes": 9845,
-        "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+        "bytes": 9926,
+        "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
       },
       "mirrors": {
         "hf_space": {
           "path": "hf_space:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         },
         "hf_artifacts": {
           "path": "hf_artifacts:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         },
         "hf_model": {
           "path": "hf_model:PROJECT_STATUS.md",
           "exists": true,
-          "bytes": 9845,
-          "sha256": "e77d3facc533bffe35586e4de6500400352c07b4ca0df5ffc523855f38faa26e"
+          "bytes": 9926,
+          "sha256": "c7dfb7a45f0c1ea435c16d93208a82da4227336e34f56a96d4afa04fce42438c"
         }
       },
       "failures": []

docs/data/omni_finetune_verified_result.json

@@ -48,28 +48,29 @@
       }
     ],
     "loss": "answer-token cross entropy over supervised JSON tokens",
-    "note": "This final Qwen3-Omni LoRA pass reused the selected 96/16/16 episode setup, trained on all exported train windows with validation monitoring, and preserved the held-out test split for final evaluation."
+    "note": "This current Qwen3-Omni LoRA result reuses the selected 96/16/16 episode setup and the v2 trained adapter, then applies the stricter label-contract prompt for held-out evaluation."
   },
   "evaluation": {
     "split": "test",
     "num_samples": 448,
     "held_out_episode_count": 14,
-    "json_validity_rate": 0.9977678571428571,
-    "action_macro_f1": 0.0024331644885523347,
+    "json_validity_rate": 1.0,
+    "action_macro_f1": 0.0021983997167007384,
     "subtask_accuracy": 0.002232142857142857,
-    "transition_accuracy": 0.9709821428571429,
-    "next_action_accuracy": 0.029017857142857144,
-    "contact_accuracy": 0.71875,
-    "object_micro_f1": 0.30160427807486634,
+    "transition_accuracy": 0.9732142857142857,
+    "next_action_accuracy": 0.03125,
+    "contact_accuracy": 0.7209821428571429,
+    "object_micro_f1": 0.30688228657389993,
     "quality_target": {
       "json_validity_rate": 0.98,
       "status": "met"
     },
-    "previous_validation_aware_json_validity_rate": 0.875
+    "previous_validation_aware_json_validity_rate": 0.875,
+    "previous_structured_json_v2_json_validity_rate": 0.9977678571428571
   },
-  "interpretation": "This is the final verified two-epoch Qwen3-Omni LoRA diagnostic result for the selected 128-episode setup. It meets the 98% JSON-validity target and improves transition, contact, and object metrics over the earlier validation-aware pilot, but action and subtask classification remain weak on held-out episodes, so this is still a baseline-quality diagnostic model rather than a strong Xperience-10M action recognizer.",
+  "interpretation": "This is the current verified Qwen3-Omni LoRA diagnostic result for the selected 128-episode setup. It reuses the same trained LoRA adapter as v2 but tightens the prompt-side label contract at evaluation time, reaching 100% JSON validity and small gains in transition, contact, next-action exact accuracy, and object micro-F1. Action and subtask classification remain weak on held-out episodes, so this is still a baseline-quality diagnostic model rather than a strong Xperience-10M action recognizer.",
   "public_package": {
-    "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full",
+    "path": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
     "audit_status": "pass",
     "contains_raw_xperience10m_data": false,
     "contains_qwen_base_weights": false,
@@ -77,9 +78,9 @@
     "adapter_weights_repo": "cy0307/ropedia-qwen3-omni-lora-128ep"
   },
   "required_next_steps": [
-    "Verify the public Hugging Face LoRA adapter repository hashes after publication.",
-    "Publish the final verified package and refreshed comparison tables to all public mirrors, then run live publication verification.",
-    "Use the full-eval predictions for error analysis focused on action/subtask confusions and unseen-label behavior.",
-    "Keep the same verified package contract for the Cosmos3 world-model branch and any future VLA/policy branches."
+    "Use the v3 strict-label predictions for action/subtask error analysis and unseen-label debugging.",
+    "Keep the existing Qwen LoRA adapter repository as the weight-bearing artifact; v3 is an evaluation/package refresh over the same adapter, not new weights.",
+    "Implement the Cosmos3-Super diffusion/action target packer and supervised loss before claiming Cosmos3 fine-tuning.",
+    "Use sharded Qwen eval for future long held-out passes to improve GPU utilization."
   ]
 }

docs/data/omni_model_comparison.json

@@ -1,6 +1,6 @@
 {
   "title": "Ropedia Xperience-10M Current Result Versions and Model Groups",
-  "generated_at_utc": "2026-06-07T09:05:41+00:00",
+  "generated_at_utc": "2026-06-07T15:34:51+00:00",
   "status": "pass",
   "version_count": 3,
   "model_group_count": 4,
@@ -313,8 +313,8 @@
       "source": "results/omni_finetune/verified_public/",
       "split": "episode/session held-out split; exact task target depends on backbone contract",
       "counts": {
-        "verified_branch_count": 5,
-        "qwen3_verified_package_count": 3,
+        "verified_branch_count": 6,
+        "qwen3_verified_package_count": 4,
         "cosmos3_verified_package_count": 2,
         "cosmos3_nano_verified_package_count": 1,
         "cosmos3_super_verified_package_count": 1
@@ -550,6 +550,58 @@
               "global_step": 712
             }
           ],
+          "is_current": false,
+          "weights_repository": "historical diagnostic package; keep separate from the final 128-episode adapter repo"
+        },
+        {
+          "id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "title": "Qwen3-Omni LoRA",
+          "status": "verified",
+          "backbone": "qwen3_omni_lora",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "structured_episode_understanding_json_qa",
+          "source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora",
+          "eval_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 1.0,
+            "action_macro_f1": 0.0021983997167007384,
+            "subtask_accuracy": 0.002232142857142857,
+            "transition_accuracy": 0.9732142857142857,
+            "next_action_accuracy": 0.03125,
+            "contact_accuracy": 0.7209821428571429,
+            "object_micro_f1": 0.30688228657389993,
+            "held_out_episode_count": 14
+          },
+          "history": [
+            {
+              "epoch": 1,
+              "train_loss": 0.41282760031950355,
+              "val_loss": 0.03288277983665466,
+              "global_step": 356
+            },
+            {
+              "epoch": 2,
+              "train_loss": 0.027745448225544075,
+              "val_loss": 0.027823254466056824,
+              "global_step": 712
+            }
+          ],
           "is_current": true,
           "weights_repository": "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep"
         }
@@ -787,6 +839,58 @@
               "global_step": 712
             }
           ],
+          "is_current": false,
+          "weights_repository": "historical diagnostic package; keep separate from the final 128-episode adapter repo"
+        },
+        {
+          "id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "title": "Qwen3-Omni LoRA",
+          "status": "verified",
+          "backbone": "qwen3_omni_lora",
+          "dataset_contract": "xperience10m_episode_json_qa_v1",
+          "training_objective": "structured_episode_understanding_json_qa",
+          "source": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json",
+          "dataset_run_id": "xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605",
+          "train_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora",
+          "eval_run_id": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+          "counts": {
+            "dataset_samples": 3808,
+            "dataset_episodes": 119,
+            "split_counts": {
+              "train": 2848,
+              "val": 512,
+              "test": 448
+            },
+            "train_samples": 2848,
+            "val_samples": 512,
+            "eval_samples": 448,
+            "held_out_episode_count": 14,
+            "num_processes": 8
+          },
+          "primary_metrics": {
+            "json_validity_rate": 1.0,
+            "action_macro_f1": 0.0021983997167007384,
+            "subtask_accuracy": 0.002232142857142857,
+            "transition_accuracy": 0.9732142857142857,
+            "next_action_accuracy": 0.03125,
+            "contact_accuracy": 0.7209821428571429,
+            "object_micro_f1": 0.30688228657389993,
+            "held_out_episode_count": 14
+          },
+          "history": [
+            {
+              "epoch": 1,
+              "train_loss": 0.41282760031950355,
+              "val_loss": 0.03288277983665466,
+              "global_step": 356
+            },
+            {
+              "epoch": 2,
+              "train_loss": 0.027745448225544075,
+              "val_loss": 0.027823254466056824,
+              "global_step": 712
+            }
+          ],
           "is_current": true,
           "weights_repository": "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep"
         }
@@ -874,6 +978,43 @@
           "interpretation": "No one-episode Cosmos3-Super adapter or fine-tuned weight run is published. The available Super result is the 128-episode held-out base-weight evaluation."
         }
       ],
+      "readiness_runs": [
+        {
+          "id": "xperience10m_cosmos3_super_training_readiness_20260607",
+          "title": "Cosmos3-Super Training Readiness Probe",
+          "scope": "selected 128-episode 96/16/16 JSON-task dataset and staged Cosmos3-Super runtime",
+          "status": "blocked_until_trainer_implemented",
+          "source": "results/omni_finetune/xperience10m_cosmos3_super_training_readiness_20260607/training_readiness.json",
+          "split": "train/val/test by selected episode/session",
+          "counts": {
+            "dataset_samples": 3808,
+            "split_counts": {
+              "test": {
+                "samples": 448,
+                "episodes": 14,
+                "actions": 189
+              },
+              "train": {
+                "samples": 2848,
+                "episodes": 89,
+                "actions": 885
+              },
+              "val": {
+                "samples": 512,
+                "episodes": 16,
+                "actions": 223
+              }
+            }
+          },
+          "primary_metrics": {
+            "diffusers_runtime_supported": true,
+            "chat_sft_supported": false,
+            "weights_updated": false
+          },
+          "weights": "none; readiness audit only, no adapter checkpoint",
+          "interpretation": "This probe confirms the staged Cosmos3-Super Diffusers/GPU runtime and the same JSON QA dataset are visible, but blocks true fine-tuning until a Cosmos-specific diffusion/action target packer and supervised loss are implemented."
+        }
+      ],
       "multi_episode_128_runs": [
         {
           "id": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
@@ -915,7 +1056,7 @@
           "weights_repository": "none for this run: staged base nv-community/Cosmos3-Super weights were evaluated through vLLM; create a separate repo only after new adapter or fine-tuned weights exist"
         }
       ],
-      "comparison_note": "Cosmos3-Super is now represented by a verified 448-window held-out Reasoner evaluation on the same JSON task as Qwen3. It uses staged base weights through vLLM, so it is a model-branch diagnostic, not a weight release."
+      "comparison_note": "Cosmos3-Super is now represented by a verified 448-window held-out Reasoner evaluation on the same JSON task as Qwen3. It uses staged base weights through vLLM, so it is a model-branch diagnostic, not a weight release. The readiness probe records why true Cosmos3-Super fine-tuning is not launched yet."
     }
   ],
   "model_group_reading_notes": [
@@ -923,10 +1064,10 @@
     "Task-head baselines have both a one-episode public-sample run and a 128-episode same-split metadata/text run.",
     "Qwen3-Omni has a one-episode sensor-adapter smoke test and separate 128-episode LoRA diagnostic packages; only the final 128-episode adapter belongs in the Qwen LoRA model repo.",
     "Cosmos3-Nano has a 128-episode future-window compatibility package.",
-    "Cosmos3-Super has a 128-episode base-weight Reasoner evaluation on the JSON task; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist."
+    "Cosmos3-Super has a 128-episode base-weight Reasoner evaluation on the JSON task plus a training-readiness probe; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist."
   ],
   "pending": [
     "Use the final Qwen3 full-eval package as the current Qwen result; older Qwen package rows remain historical diagnostics for comparison.",
-    "Promote Cosmos3 from Nano compatibility and Super base-weight evaluation to true fine-tuning only after a dedicated Cosmos adapter/diffusion training path produces new weights."
+    "Promote Cosmos3 from Nano compatibility and Super base-weight evaluation to true fine-tuning only after a dedicated Cosmos diffusion/action target packer and supervised loss produce new weights."
   ]
 }

docs/data/project_status.json

@@ -214,13 +214,13 @@
       "status": "final_verified_diagnostic_result_json_target_met",
       "evidence": [
         "docs/data/omni_finetune_verified_result.json",
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/",
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/",
         "https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep",
         "scripts/omni/package_verified_omni_result.py",
         "scripts/omni/audit_verified_omni_package.py",
         "scripts/omni/analyze_qwen3_omni_errors.py"
       ],
-      "readout": "The selected 96/16/16 episode split produced a final public-safe held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs, validation/audit summaries, and a public LoRA adapter repo. JSON validity is 99.78%, meeting the 98% target; transition accuracy is 97.10%, contact accuracy is 71.88%, object micro-F1 is 30.16%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim."
+      "readout": "The selected 96/16/16 episode split now has a v3 strict-label public-safe held-out package with 3,808 exported windows, 512 validation windows, 448 test predictions, two training epochs reused from the same LoRA adapter, validation/audit summaries, and a public LoRA adapter repo. JSON validity is 100.00%, meeting the 98% target; transition accuracy is 97.32%, contact accuracy is 72.10%, object micro-F1 is 30.69%, and action/subtask metrics remain weak, so it is still a diagnostic baseline rather than a strong model-quality claim."
     },
     {
       "area": "Cosmos3-Nano future-window branch",

docs/data/publication_audit.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:41:02+00:00",
+  "generated_at_utc": "2026-06-07T15:49:07+00:00",
   "checks": [
     {
       "name": "required_publication_assets_present",
@@ -182,8 +182,8 @@
     "github_repo": {
       "root": "repo",
       "exists": true,
-      "file_count": 618,
-      "text_file_count": 523,
+      "file_count": 680,
+      "text_file_count": 577,
       "largest_file": {
         "path": "tmp/omni_128_dataset_fetch/dataset.jsonl",
         "bytes": 582271586
@@ -193,8 +193,8 @@
     "hf_space_bundle": {
       "root": "hf_publish/space",
       "exists": true,
-      "file_count": 519,
-      "text_file_count": 425,
+      "file_count": 582,
+      "text_file_count": 480,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -204,8 +204,8 @@
     "hf_artifact_bundle": {
       "root": "hf_publish/artifacts",
       "exists": true,
-      "file_count": 695,
-      "text_file_count": 577,
+      "file_count": 757,
+      "text_file_count": 631,
       "largest_file": {
         "path": "results/episode_task_suite/modality_reconstruction/predictions.npz",
         "bytes": 55702978
@@ -215,8 +215,8 @@
     "hf_model_bundle": {
       "root": "hf_publish/model",
       "exists": true,
-      "file_count": 883,
-      "text_file_count": 730,
+      "file_count": 945,
+      "text_file_count": 784,
       "largest_file": {
         "path": "pytorch_model.bin",
         "bytes": 93495480

docs/data/reproducibility_matrix.json

@@ -82,7 +82,7 @@
       "status": "verified_final_diagnostic_result_not_publicly_rerunnable_without_gated_data",
       "command": "scripts/omni/build_qwen3_omni_dataset.py and scripts/omni/train_qwen3_omni_lora.py on the selected gated episodes",
       "expected": "verified final diagnostic LoRA package with 3,808 exported windows, 2,848 train windows, and 448 held-out test predictions",
-      "boundary": "the public package records metrics and manifests, but rerunning requires gated Xperience-10M episode access and base-model weights; current JSON validity is 99.78%, meeting the 98% target, while action/subtask metrics remain weak"
+      "boundary": "the public package records metrics and manifests, but rerunning requires gated Xperience-10M episode access and base-model weights; current strict-label JSON validity is 100.00%, meeting the 98% target, while action/subtask metrics remain weak"
     }
   ]
 }

docs/data/scope_claims_audit.json

@@ -1,15 +1,15 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T00:00:35+00:00",
+  "generated_at_utc": "2026-06-07T15:47:31+00:00",
   "summary": {
     "qwen3_omni_verified_diagnostic_pilot": true,
     "dataset_manifest_num_episodes": 119,
     "dataset_manifest_num_samples": 3808,
     "training_metadata_num_train_samples": 2848,
     "eval_num_samples": 448,
-    "eval_json_validity_rate": 0.9977678571428571,
+    "eval_json_validity_rate": 1.0,
     "quality_target_met": true,
-    "historical_identifier_count": 189,
+    "historical_identifier_count": 1545,
     "public_32_episode_status_file_count": 1,
     "failure_count": 0
   },
@@ -35,7 +35,7 @@
       "status": "pass",
       "detail": "episodes=119, samples=3808, split_counts={'train': 2848, 'val': 512, 'test': 448}",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/dataset/dataset_manifest.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/dataset_manifest.json"
       ]
     },
     {
@@ -43,15 +43,15 @@
       "status": "pass",
       "detail": "train=2848, val=512, processes=8",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/training/training_metadata.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/training/training_metadata.json"
       ]
     },
     {
       "name": "verified_package_eval_records_real_held_out_metrics",
       "status": "pass",
-      "detail": "samples=448, split=test, held_out=14, json_validity=0.9977678571428571",
+      "detail": "samples=448, split=test, held_out=14, json_validity=1.0",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/eval/metrics.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval/metrics.json"
       ]
     },
     {
@@ -59,7 +59,7 @@
       "status": "pass",
       "detail": "audit_status=pass, issues=0",
       "evidence": [
-        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/package_audit.json"
+        "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/package_audit.json"
       ]
     },
     {
@@ -84,7 +84,7 @@
     {
       "name": "historical_32ep_identifiers_are_confined_to_readiness_artifacts",
       "status": "pass",
-      "detail": "historical identifiers found in result provenance files=189",
+      "detail": "historical identifiers found in result provenance files=1545",
       "evidence": [
         "results/omni_finetune/"
       ]
@@ -424,6 +424,6 @@
       "example": "{\"id\": \"xperience-10m-sample:qa:53\", \"episode_id\": \"xperience-10m-sample\", \"split\": \"train\", \"target\": \"episode_qa\", \"prompt_type\": \"json_episode_understanding\", \"center_window\": {\"start_frame\": 1060, \"end_frame\": 1079, \"num_frames\": 20}, \"media\": {\"video_path"
     }
   ],
-  "historical_identifier_total_count": 189,
+  "historical_identifier_total_count": 1545,
   "failures": []
 }

docs/data/task_surface_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-06T23:27:06+00:00",
+  "generated_at_utc": "2026-06-07T15:47:30+00:00",
   "summary": {
     "task_count": 12,
     "expected_task_count": 12,
@@ -64,15 +64,15 @@
       "observed": "timeline_action"
     },
     {
-      "name": "timeline_action: public_field_plain_goal_is_human_readable",
+      "name": "timeline_action: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Look at one short multimodal window and name what action is happening now.",
+      "value": "20-frame multimodal window",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_card_blurb_is_human_readable",
+      "name": "timeline_action: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Recognize the current manipulation action from synchronized visual, motion, inertial, pose, and annotation context.",
+      "value": "Egocentric Action Recognition",
       "raw_hits": []
     },
     {
@@ -82,27 +82,27 @@
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_input_short_is_human_readable",
+      "name": "timeline_action: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "20-frame multimodal window",
+      "value": "Recognize the current manipulation action from synchronized visual, motion, inertial, pose, and annotation context.",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_output_short_is_human_readable",
+      "name": "timeline_action: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "current action class",
+      "value": "window features -> action label builder -> classifier",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_process_short_is_human_readable",
+      "name": "timeline_action: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "window features -> action label builder -> classifier",
+      "value": "current action class",
       "raw_hits": []
     },
     {
-      "name": "timeline_action: public_field_research_name_is_human_readable",
+      "name": "timeline_action: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Egocentric Action Recognition",
+      "value": "Look at one short multimodal window and name what action is happening now.",
       "raw_hits": []
     },
     {
@@ -184,15 +184,15 @@
       "observed": "timeline_subtask"
     },
     {
-      "name": "timeline_subtask: public_field_plain_goal_is_human_readable",
+      "name": "timeline_subtask: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict the higher-level task stage for the current window.",
+      "value": "20-frame multimodal window",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_card_blurb_is_human_readable",
+      "name": "timeline_subtask: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Recognize the broader activity stage so fine actions become a readable procedure timeline.",
+      "value": "Temporal Subtask Recognition",
       "raw_hits": []
     },
     {
@@ -202,27 +202,27 @@
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_input_short_is_human_readable",
+      "name": "timeline_subtask: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "20-frame multimodal window",
+      "value": "Recognize the broader activity stage so fine actions become a readable procedure timeline.",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_output_short_is_human_readable",
+      "name": "timeline_subtask: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "current procedure step",
+      "value": "window features -> subtask label builder -> classifier",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_process_short_is_human_readable",
+      "name": "timeline_subtask: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "window features -> subtask label builder -> classifier",
+      "value": "current procedure step",
       "raw_hits": []
     },
     {
-      "name": "timeline_subtask: public_field_research_name_is_human_readable",
+      "name": "timeline_subtask: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Subtask Recognition",
+      "value": "Predict the higher-level task stage for the current window.",
       "raw_hits": []
     },
     {
@@ -304,15 +304,15 @@
       "observed": "transition_detection"
     },
     {
-      "name": "transition_detection: public_field_plain_goal_is_human_readable",
+      "name": "transition_detection: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Detect whether the current window is near a boundary between actions.",
+      "value": "current window with boundary target",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_card_blurb_is_human_readable",
+      "name": "transition_detection: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Detect the local moment where the episode changes from one action segment to the next.",
+      "value": "Temporal Action Segmentation",
       "raw_hits": []
     },
     {
@@ -322,27 +322,27 @@
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_input_short_is_human_readable",
+      "name": "transition_detection: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current window with boundary target",
+      "value": "Detect the local moment where the episode changes from one action segment to the next.",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_output_short_is_human_readable",
+      "name": "transition_detection: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "boundary or steady",
+      "value": "action changes -> boundary labels -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_process_short_is_human_readable",
+      "name": "transition_detection: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "action changes -> boundary labels -> binary classifier",
+      "value": "boundary or steady",
       "raw_hits": []
     },
     {
-      "name": "transition_detection: public_field_research_name_is_human_readable",
+      "name": "transition_detection: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Action Segmentation",
+      "value": "Detect whether the current window is near a boundary between actions.",
       "raw_hits": []
     },
     {
@@ -422,15 +422,15 @@
       "observed": "next_action"
     },
     {
-      "name": "next_action: public_field_plain_goal_is_human_readable",
+      "name": "next_action: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Use the current window to guess the action that will happen shortly after it.",
+      "value": "current window at time t",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_card_blurb_is_human_readable",
+      "name": "next_action: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Forecast the near-future action from the current observations only.",
+      "value": "Short-Horizon Intention Prediction",
       "raw_hits": []
     },
     {
@@ -440,27 +440,27 @@
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_input_short_is_human_readable",
+      "name": "next_action: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current window at time t",
+      "value": "Forecast the near-future action from the current observations only.",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_output_short_is_human_readable",
+      "name": "next_action: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "action at t+20 frames",
+      "value": "current features -> future label shift -> classifier",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_process_short_is_human_readable",
+      "name": "next_action: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "current features -> future label shift -> classifier",
+      "value": "action at t+20 frames",
       "raw_hits": []
     },
     {
-      "name": "next_action: public_field_research_name_is_human_readable",
+      "name": "next_action: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Short-Horizon Intention Prediction",
+      "value": "Use the current window to guess the action that will happen shortly after it.",
       "raw_hits": []
     },
     {
@@ -540,15 +540,15 @@
       "observed": "hand_trajectory_forecast"
     },
     {
-      "name": "hand_trajectory_forecast: public_field_plain_goal_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict where the hands will move over the next few frames.",
+      "value": "current multimodal window",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_card_blurb_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict the future 3D left/right hand path from the current multimodal state.",
+      "value": "3D Hand Motion Forecasting",
       "raw_hits": []
     },
     {
@@ -558,27 +558,27 @@
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_input_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "current multimodal window",
+      "value": "Predict the future 3D left/right hand path from the current multimodal state.",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_output_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "future hand-joint trajectory",
+      "value": "current features -> future mocap target -> regression head",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_process_short_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "current features -> future mocap target -> regression head",
+      "value": "future hand-joint trajectory",
       "raw_hits": []
     },
     {
-      "name": "hand_trajectory_forecast: public_field_research_name_is_human_readable",
+      "name": "hand_trajectory_forecast: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "3D Hand Motion Forecasting",
+      "value": "Predict where the hands will move over the next few frames.",
       "raw_hits": []
     },
     {
@@ -658,15 +658,15 @@
       "observed": "contact_prediction"
     },
     {
-      "name": "contact_prediction: public_field_plain_goal_is_human_readable",
+      "name": "contact_prediction: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict whether the body or hand is in contact with something.",
+      "value": "non-contact, non-caption features",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_card_blurb_is_human_readable",
+      "name": "contact_prediction: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict whether body or hand contact with the scene is occurring without leaking contact labels.",
+      "value": "Human-Object Contact Prediction",
       "raw_hits": []
     },
     {
@@ -676,27 +676,27 @@
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_input_short_is_human_readable",
+      "name": "contact_prediction: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "non-contact, non-caption features",
+      "value": "Predict whether body or hand contact with the scene is occurring without leaking contact labels.",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_output_short_is_human_readable",
+      "name": "contact_prediction: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "contact or no contact",
+      "value": "feature filter -> contact target -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_process_short_is_human_readable",
+      "name": "contact_prediction: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "feature filter -> contact target -> binary classifier",
+      "value": "contact or no contact",
       "raw_hits": []
     },
     {
-      "name": "contact_prediction: public_field_research_name_is_human_readable",
+      "name": "contact_prediction: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Human-Object Contact Prediction",
+      "value": "Predict whether the body or hand is in contact with something.",
       "raw_hits": []
     },
     {
@@ -774,15 +774,15 @@
       "observed": "object_relevance"
     },
     {
-      "name": "object_relevance: public_field_plain_goal_is_human_readable",
+      "name": "object_relevance: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict which objects matter in the current window.",
+      "value": "non-caption multimodal features",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_card_blurb_is_human_readable",
+      "name": "object_relevance: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Infer which objects are relevant to the current manipulation window from non-caption features.",
+      "value": "Object-Centric Interaction Recognition",
       "raw_hits": []
     },
     {
@@ -792,27 +792,27 @@
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_input_short_is_human_readable",
+      "name": "object_relevance: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "non-caption multimodal features",
+      "value": "Infer which objects are relevant to the current manipulation window from non-caption features.",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_output_short_is_human_readable",
+      "name": "object_relevance: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "relevant object set",
+      "value": "object vocabulary -> multi-hot labels -> sigmoid heads",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_process_short_is_human_readable",
+      "name": "object_relevance: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "object vocabulary -> multi-hot labels -> sigmoid heads",
+      "value": "relevant object set",
       "raw_hits": []
     },
     {
-      "name": "object_relevance: public_field_research_name_is_human_readable",
+      "name": "object_relevance: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Object-Centric Interaction Recognition",
+      "value": "Predict which objects matter in the current window.",
       "raw_hits": []
     },
     {
@@ -892,15 +892,15 @@
       "observed": "caption_grounding"
     },
     {
-      "name": "caption_grounding: public_field_plain_goal_is_human_readable",
+      "name": "caption_grounding: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Given a text-like query from annotation, find the matching time window.",
+      "value": "text-like query and candidate windows",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_card_blurb_is_human_readable",
+      "name": "caption_grounding: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Retrieve the matching time window for an annotation-derived text query.",
+      "value": "Language-to-Moment Grounding",
       "raw_hits": []
     },
     {
@@ -910,27 +910,27 @@
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_input_short_is_human_readable",
+      "name": "caption_grounding: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "text-like query and candidate windows",
+      "value": "Retrieve the matching time window for an annotation-derived text query.",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_output_short_is_human_readable",
+      "name": "caption_grounding: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "ranked matching moments",
+      "value": "query features -> candidate index -> cosine ranker",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_process_short_is_human_readable",
+      "name": "caption_grounding: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "query features -> candidate index -> cosine ranker",
+      "value": "ranked matching moments",
       "raw_hits": []
     },
     {
-      "name": "caption_grounding: public_field_research_name_is_human_readable",
+      "name": "caption_grounding: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Language-to-Moment Grounding",
+      "value": "Given a text-like query from annotation, find the matching time window.",
       "raw_hits": []
     },
     {
@@ -1008,15 +1008,15 @@
       "observed": "cross_modal_retrieval"
     },
     {
-      "name": "cross_modal_retrieval: public_field_plain_goal_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Use one group of modalities to retrieve the matching window from another group.",
+      "value": "motion/IMU/pose query; depth/video candidates",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_card_blurb_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Use motion, IMU, and camera-pose signals to retrieve the matching depth/video window.",
+      "value": "Multimodal Representation Retrieval",
       "raw_hits": []
     },
     {
@@ -1026,27 +1026,27 @@
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_input_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion/IMU/pose query; depth/video candidates",
+      "value": "Use motion, IMU, and camera-pose signals to retrieve the matching depth/video window.",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_output_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "ranked visual windows",
+      "value": "modality split -> projection -> nearest-neighbor ranker",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_process_short_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "modality split -> projection -> nearest-neighbor ranker",
+      "value": "ranked visual windows",
       "raw_hits": []
     },
     {
-      "name": "cross_modal_retrieval: public_field_research_name_is_human_readable",
+      "name": "cross_modal_retrieval: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Multimodal Representation Retrieval",
+      "value": "Use one group of modalities to retrieve the matching window from another group.",
       "raw_hits": []
     },
     {
@@ -1126,15 +1126,15 @@
       "observed": "modality_reconstruction"
     },
     {
-      "name": "modality_reconstruction: public_field_plain_goal_is_human_readable",
+      "name": "modality_reconstruction: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Predict one modality feature block from other modality blocks.",
+      "value": "motion, IMU, and camera/pose features",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_card_blurb_is_human_readable",
+      "name": "modality_reconstruction: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Predict compressed depth/video feature vectors from motion, IMU, and camera-pose features.",
+      "value": "Modality Feature Reconstruction",
       "raw_hits": []
     },
     {
@@ -1144,27 +1144,27 @@
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_input_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion, IMU, and camera/pose features",
+      "value": "Predict compressed depth/video feature vectors from motion, IMU, and camera-pose features.",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_output_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "reconstructed depth/video vector",
+      "value": "source-target split -> scaler -> regression head",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_process_short_is_human_readable",
+      "name": "modality_reconstruction: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "source-target split -> scaler -> regression head",
+      "value": "reconstructed depth/video vector",
       "raw_hits": []
     },
     {
-      "name": "modality_reconstruction: public_field_research_name_is_human_readable",
+      "name": "modality_reconstruction: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Modality Feature Reconstruction",
+      "value": "Predict one modality feature block from other modality blocks.",
       "raw_hits": []
     },
     {
@@ -1244,15 +1244,15 @@
       "observed": "temporal_order"
     },
     {
-      "name": "temporal_order: public_field_plain_goal_is_human_readable",
+      "name": "temporal_order: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Tell whether two nearby windows are in the correct time order.",
+      "value": "two adjacent windows plus difference vector",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_card_blurb_is_human_readable",
+      "name": "temporal_order: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Tell whether two neighboring windows are in chronological order or reversed.",
+      "value": "Temporal Order Verification",
       "raw_hits": []
     },
     {
@@ -1262,27 +1262,27 @@
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_input_short_is_human_readable",
+      "name": "temporal_order: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "two adjacent windows plus difference vector",
+      "value": "Tell whether two neighboring windows are in chronological order or reversed.",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_output_short_is_human_readable",
+      "name": "temporal_order: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "correct or reversed",
+      "value": "pair builder -> feature combiner -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_process_short_is_human_readable",
+      "name": "temporal_order: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "pair builder -> feature combiner -> binary classifier",
+      "value": "correct or reversed",
       "raw_hits": []
     },
     {
-      "name": "temporal_order: public_field_research_name_is_human_readable",
+      "name": "temporal_order: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Temporal Order Verification",
+      "value": "Tell whether two nearby windows are in the correct time order.",
       "raw_hits": []
     },
     {
@@ -1360,15 +1360,15 @@
       "observed": "misalignment_detection"
     },
     {
-      "name": "misalignment_detection: public_field_plain_goal_is_human_readable",
+      "name": "misalignment_detection: public_field_input_short_is_human_readable",
       "status": "pass",
-      "value": "Detect when modalities that should match are shifted out of sync.",
+      "value": "motion-side and visual/depth-side feature groups",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_card_blurb_is_human_readable",
+      "name": "misalignment_detection: public_field_research_name_is_human_readable",
       "status": "pass",
-      "value": "Detect whether motion and visual/depth streams have been artificially shifted out of sync.",
+      "value": "Cross-Modal Misalignment Detection",
       "raw_hits": []
     },
     {
@@ -1378,27 +1378,27 @@
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_input_short_is_human_readable",
+      "name": "misalignment_detection: public_field_card_blurb_is_human_readable",
       "status": "pass",
-      "value": "motion-side and visual/depth-side feature groups",
+      "value": "Detect whether motion and visual/depth streams have been artificially shifted out of sync.",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_output_short_is_human_readable",
+      "name": "misalignment_detection: public_field_process_short_is_human_readable",
       "status": "pass",
-      "value": "aligned or shifted",
+      "value": "aligned/shifted pairs -> feature combiner -> binary classifier",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_process_short_is_human_readable",
+      "name": "misalignment_detection: public_field_output_short_is_human_readable",
       "status": "pass",
-      "value": "aligned/shifted pairs -> feature combiner -> binary classifier",
+      "value": "aligned or shifted",
       "raw_hits": []
     },
     {
-      "name": "misalignment_detection: public_field_research_name_is_human_readable",
+      "name": "misalignment_detection: public_field_plain_goal_is_human_readable",
       "status": "pass",
-      "value": "Cross-Modal Misalignment Detection",
+      "value": "Detect when modalities that should match are shifted out of sync.",
       "raw_hits": []
     },
     {

docs/data/website_integrity.json

@@ -1,6 +1,6 @@
 {
   "status": "pass",
-  "generated_at_utc": "2026-06-07T09:40:59+00:00",
+  "generated_at_utc": "2026-06-07T15:47:32+00:00",
   "docs_root": "docs",
   "site_base": "/ropedia-xperience-10m-task-suite/",
   "summary": {
@@ -75,7 +75,7 @@
       "status": "pass",
       "reason": "The project overview should appear before the deeper progress ledger.",
       "overview_index": 67412,
-      "evidence_index": 90476
+      "evidence_index": 90477
     },
     {
       "name": "project_status_links_json",
@@ -153,8 +153,8 @@
       "status": "pass",
       "reason": "The evaluation protocol should appear before the deeper evidence ledger.",
       "overview_index": 67412,
-      "protocol_index": 87159,
-      "evidence_index": 90476
+      "protocol_index": 87160,
+      "evidence_index": 90477
     },
     {
       "name": "evaluation_protocol_links_json",
@@ -252,7 +252,7 @@
     },
     {
       "path": "data/artifact_index.json",
-      "bytes": 60162,
+      "bytes": 67401,
       "top_level_type": "dict"
     },
     {
@@ -292,7 +292,7 @@
     },
     {
       "path": "data/mirror_parity.json",
-      "bytes": 257049,
+      "bytes": 410374,
       "top_level_type": "dict"
     },
     {
@@ -302,12 +302,12 @@
     },
     {
       "path": "data/omni_finetune_verified_result.json",
-      "bytes": 3483,
+      "bytes": 3628,
       "top_level_type": "dict"
     },
     {
       "path": "data/omni_model_comparison.json",
-      "bytes": 42015,
+      "bytes": 48296,
       "top_level_type": "dict"
     },
     {
@@ -327,7 +327,7 @@
     },
     {
       "path": "data/project_status.json",
-      "bytes": 16400,
+      "bytes": 16455,
       "top_level_type": "dict"
     },
     {
@@ -352,7 +352,7 @@
     },
     {
       "path": "data/reproducibility_matrix.json",
-      "bytes": 5280,
+      "bytes": 5294,
       "top_level_type": "dict"
     },
     {
@@ -382,7 +382,7 @@
     },
     {
       "path": "data/scope_claims_audit.json",
-      "bytes": 21234,
+      "bytes": 21251,
       "top_level_type": "dict"
     },
     {

docs/index.html

@@ -2328,7 +2328,7 @@
             <div class="snapshot-meta">
               <span>split <strong>96 / 16 / 16</strong></span>
               <span>test windows <strong>448</strong></span>
-              <span>JSON validity <strong>99.78%</strong></span>
+              <span>JSON validity <strong>100.00%</strong></span>
             </div>
           </article>
           <article class="snapshot-card gated">
@@ -2707,7 +2707,7 @@
           <article class="artifact primary-artifact"><div><h3>Official dataset</h3><p>Xperience-10M is a gated large-scale egocentric multimodal dataset for embodied AI, robotics, spatial intelligence, and world modeling.</p></div><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m">official HF dataset</a></article>
           <article class="artifact"><h3>Public sample</h3><p>The current task suite is built from one public sample episode, not from the entire gated dataset.</p><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m-sample">sample dataset</a></article>
           <article class="artifact"><h3>Modalities</h3><p>The sample exposes synchronized video, audio, depth, pose/SLAM, motion capture, inertial signals, calibration, and language annotations.</p><a href="data/modality_atlas.json">modality atlas</a></article>
-          <article class="artifact"><h3>Multi-episode pilot</h3><p>The selected 128-episode Qwen3-Omni LoRA final diagnostic result is verified with 448 held-out test predictions and 99.78% JSON validity. Action/subtask metrics are still weak, so this remains a baseline for error analysis.</p><a href="https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep">LoRA adapter</a></article>
+          <article class="artifact"><h3>Multi-episode pilot</h3><p>The selected 128-episode Qwen3-Omni LoRA strict-label v3 diagnostic result is verified with 448 held-out test predictions and 100.00% JSON validity. Action/subtask metrics are still weak, so this remains a baseline for error analysis.</p><a href="https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep">LoRA adapter</a></article>
           <article class="artifact"><h3>Data boundary</h3><p>Raw MP4, HDF5, RRD files, private gated data, and full Qwen weights are not redistributed in this project.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/DATA_NOTICE.md">data notice</a></article>
           <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 aligned windows, 8,546-dimensional task inputs, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
           <article class="artifact"><h3>Covered now</h3><p>Action/subtask labels, next-action prediction, temporal diagnostics, hand trajectory, contact, object relevance, caption grounding, retrieval, reconstruction, and misalignment.</p><a href="data/summary_metrics.json">summary metrics</a></article>

index.html

@@ -2328,7 +2328,7 @@
             <div class="snapshot-meta">
               <span>split <strong>96 / 16 / 16</strong></span>
               <span>test windows <strong>448</strong></span>
-              <span>JSON validity <strong>99.78%</strong></span>
+              <span>JSON validity <strong>100.00%</strong></span>
             </div>
           </article>
           <article class="snapshot-card gated">
@@ -2707,7 +2707,7 @@
           <article class="artifact primary-artifact"><div><h3>Official dataset</h3><p>Xperience-10M is a gated large-scale egocentric multimodal dataset for embodied AI, robotics, spatial intelligence, and world modeling.</p></div><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m">official HF dataset</a></article>
           <article class="artifact"><h3>Public sample</h3><p>The current task suite is built from one public sample episode, not from the entire gated dataset.</p><a href="https://huggingface.co/datasets/ropedia-ai/xperience-10m-sample">sample dataset</a></article>
           <article class="artifact"><h3>Modalities</h3><p>The sample exposes synchronized video, audio, depth, pose/SLAM, motion capture, inertial signals, calibration, and language annotations.</p><a href="data/modality_atlas.json">modality atlas</a></article>
-          <article class="artifact"><h3>Multi-episode pilot</h3><p>The selected 128-episode Qwen3-Omni LoRA final diagnostic result is verified with 448 held-out test predictions and 99.78% JSON validity. Action/subtask metrics are still weak, so this remains a baseline for error analysis.</p><a href="https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep">LoRA adapter</a></article>
+          <article class="artifact"><h3>Multi-episode pilot</h3><p>The selected 128-episode Qwen3-Omni LoRA strict-label v3 diagnostic result is verified with 448 held-out test predictions and 100.00% JSON validity. Action/subtask metrics are still weak, so this remains a baseline for error analysis.</p><a href="https://huggingface.co/cy0307/ropedia-qwen3-omni-lora-128ep">LoRA adapter</a></article>
           <article class="artifact"><h3>Data boundary</h3><p>Raw MP4, HDF5, RRD files, private gated data, and full Qwen weights are not redistributed in this project.</p><a href="https://github.com/ChaoYue0307/ropedia-xperience-10m-task-suite/blob/main/DATA_NOTICE.md">data notice</a></article>
           <article class="artifact"><h3>Current project subset</h3><p>One public sample episode, 5,821 frames, 1,161 aligned windows, 8,546-dimensional task inputs, and no raw-data redistribution.</p><a href="data/modality_atlas.json">modality atlas</a></article>
           <article class="artifact"><h3>Covered now</h3><p>Action/subtask labels, next-action prediction, temporal diagnostics, hand trajectory, contact, object relevance, caption grounding, retrieval, reconstruction, and misalignment.</p><a href="data/summary_metrics.json">summary metrics</a></article>

results/omni_finetune/HF_UPLOAD.md

@@ -16,7 +16,7 @@ Prepare the upload directory from the completed adapter and verified summary:
 ```bash
 python3 scripts/omni/prepare_qwen3_lora_hf_package.py \
   --adapter-dir checkpoints/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora/adapter_lora \
-  --verified-summary results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/verified_result_summary.json \
+  --verified-summary results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json \
   --output-dir results/omni_finetune/hf_upload_qwen3_128ep_full \
   --repo-id cy0307/ropedia-qwen3-omni-lora-128ep
 ```

results/omni_finetune/OMNI_MODEL_COMPARISON.md

@@ -1,6 +1,6 @@
 # Omni Model Comparison
 
-Generated: `2026-06-07T09:05:41+00:00`
+Generated: `2026-06-07T15:34:51+00:00`
 
 Compare only rows with the same scope and target. Single-episode raw-feature metrics, 128-episode metadata baselines, Qwen3 structured JSON metrics, and the two Cosmos3 targets answer different questions: Nano future-window retrieval versus Super structured JSON Reasoner evaluation.
 
@@ -24,7 +24,7 @@ Read the three rows this way:
 - Task-head baselines have both a one-episode public-sample run and a 128-episode same-split metadata/text run.
 - Qwen3-Omni has a one-episode sensor-adapter smoke test and separate 128-episode LoRA diagnostic packages; only the final 128-episode adapter belongs in the Qwen LoRA model repo.
 - Cosmos3-Nano has a 128-episode future-window compatibility package.
-- Cosmos3-Super has a 128-episode base-weight Reasoner evaluation on the JSON task; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist.
+- Cosmos3-Super has a 128-episode base-weight Reasoner evaluation on the JSON task plus a training-readiness probe; create a separate Cosmos model repo only after real Cosmos adapter/fine-tuned weights exist.
 
 ### Minimal and Neural Task Heads
 
@@ -48,7 +48,8 @@ The one-episode Qwen entry is only a sensor-adapter smoke test with Qwen3 weight
 | 1 episode | verified_smoke | Qwen3-Omni Sensor-Adapter Smoke | 1 episodes, 59 windows/samples | accuracy=0.0000, macro_f1=0.0000 | `results/omni_exploration/qwen3_adapter_smoke/metrics.json` |
 | 128 episode | verified | Qwen3-Omni LoRA | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=0.8750, action_macro_f1=0.0027, transition_accuracy=0.8504, contact_accuracy=0.6451 | `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605_eval/verified_result_summary.json` |
 | 128 episode | verified | Qwen3-Omni LoRA | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=0.8527, action_macro_f1=0.0021, transition_accuracy=0.8281, contact_accuracy=0.6518 | `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_fullsplit_fast8gpu_lora_fsdp_full_train_noval_tail_logits_fullstatesave_v6_eval_test_full/verified_result_summary.json` |
-| 128 episode | verified current | Qwen3-Omni LoRA | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=0.9978, action_macro_f1=0.0024, transition_accuracy=0.9710, contact_accuracy=0.7188 | `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/verified_result_summary.json` |
+| 128 episode | verified | Qwen3-Omni LoRA | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=0.9978, action_macro_f1=0.0024, transition_accuracy=0.9710, contact_accuracy=0.7188 | `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/verified_result_summary.json` |
+| 128 episode | verified current | Qwen3-Omni LoRA | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=1.0000, action_macro_f1=0.0022, transition_accuracy=0.9732, contact_accuracy=0.7210 | `results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/verified_result_summary.json` |
 
 ### Cosmos3-Nano Future-Window World Model
 
@@ -63,13 +64,14 @@ The current 128-episode Cosmos result is a public-safe future-window compatibili
 
 ### Cosmos3-Super Reasoner
 
-Cosmos3-Super is now represented by a verified 448-window held-out Reasoner evaluation on the same JSON task as Qwen3. It uses staged base weights through vLLM, so it is a model-branch diagnostic, not a weight release.
+Cosmos3-Super is now represented by a verified 448-window held-out Reasoner evaluation on the same JSON task as Qwen3. It uses staged base weights through vLLM, so it is a model-branch diagnostic, not a weight release. The readiness probe records why true Cosmos3-Super fine-tuning is not launched yet.
 
 - Weight repo policy: none for this run; staged base weights only, no new fine-tuned weights
 
 | scope | status | run | counts | metrics | source |
 | --- | --- | --- | --- | --- | --- |
 | 1 episode | not_run | Cosmos3-Super One-Episode Fine-Tune |  |  |  |
+| readiness | blocked_until_trainer_implemented | Cosmos3-Super Training Readiness Probe | 3808 windows/samples | diffusers_runtime_supported=True, chat_sft_supported=False, weights_updated=False | `results/omni_finetune/xperience10m_cosmos3_super_training_readiness_20260607/training_readiness.json` |
 | 128 episode | verified current | Cosmos3-Super Reasoner | 119 episodes, 3808 windows/samples, 448 eval | json_validity_rate=0.5112, action_macro_f1=0.0008, transition_accuracy=0.3683, contact_accuracy=0.3214 | `results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/verified_result_summary.json` |
 
 ## 128-Episode Task Baselines
@@ -98,8 +100,9 @@ Cosmos3-Super is now represented by a verified 448-window held-out Reasoner eval
 | Qwen3-Omni LoRA | `qwen3_omni_lora` | 448 | 14 | json_validity_rate=0.8750, action_macro_f1=0.0027, transition_accuracy=0.8504, contact_accuracy=0.6451 |
 | Qwen3-Omni LoRA | `qwen3_omni_lora` | 448 | 14 | json_validity_rate=0.8527, action_macro_f1=0.0021, transition_accuracy=0.8281, contact_accuracy=0.6518 |
 | Qwen3-Omni LoRA | `qwen3_omni_lora` | 448 | 14 | json_validity_rate=0.9978, action_macro_f1=0.0024, transition_accuracy=0.9710, contact_accuracy=0.7188 |
+| Qwen3-Omni LoRA | `qwen3_omni_lora` | 448 | 14 | json_validity_rate=1.0000, action_macro_f1=0.0022, transition_accuracy=0.9732, contact_accuracy=0.7210 |
 
 ## Pending
 
 - Use the final Qwen3 full-eval package as the current Qwen result; older Qwen package rows remain historical diagnostics for comparison.
-- Promote Cosmos3 from Nano compatibility and Super base-weight evaluation to true fine-tuning only after a dedicated Cosmos adapter/diffusion training path produces new weights.
+- Promote Cosmos3 from Nano compatibility and Super base-weight evaluation to true fine-tuning only after a dedicated Cosmos diffusion/action target packer and supervised loss produce new weights.

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/ERROR_ANALYSIS.md

@@ -0,0 +1,78 @@
+# Cosmos3-Super Reasoner base-weight Held-Out Error Analysis
+
+This report is computed from the verified public package predictions. It contains only derived metrics and sanitized examples.
+
+## Overall
+
+- Prediction rows: `448`
+- JSON validity from `metrics.json`: `0.5112`
+- Parsed prediction rate from public rows: `0.5112`
+- Action exact rate: `0.0089`
+- Subtask exact rate: `0.0000`
+- Contact exact rate: `0.3214`
+- Object F1: `0.1370`
+
+## Weakest Episode Groups
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| 9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2 | 32 | 0.0938 | 0.0000 | 0.0325 |
+| 5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6 | 32 | 0.2500 | 0.0000 | 0.0000 |
+| b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3 | 32 | 0.2812 | 0.0000 | 0.1401 |
+| b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2 | 32 | 0.2812 | 0.0312 | 0.1439 |
+| 877779cd-25f3-4293-a3c4-39067dd9558c__ep4 | 32 | 0.3125 | 0.0000 | 0.2182 |
+| ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2 | 32 | 0.4375 | 0.0000 | 0.0000 |
+| 1796b943-caad-43c6-b9bd-80b8d601f37d__ep1 | 32 | 0.4688 | 0.0000 | 0.0578 |
+| ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2 | 32 | 0.5312 | 0.0000 | 0.1728 |
+
+## Action Families
+
+| group | samples | parsed_prediction_rate | action_exact_rate | subtask_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- | --- |
+| small_object_sorting | 87 | 0.3218 | 0.0000 | 0.0000 | 0.1299 |
+| paper_cardboard_craft | 142 | 0.3239 | 0.0141 | 0.0000 | 0.0840 |
+| other | 94 | 0.5106 | 0.0000 | 0.0000 | 0.0900 |
+| food_kitchen | 5 | 0.6000 | 0.0000 | 0.0000 | 0.0000 |
+| locomotion | 23 | 0.6957 | 0.0000 | 0.0000 | 0.0348 |
+| cleaning | 8 | 0.7500 | 0.0000 | 0.0000 | 0.0000 |
+| phone_use | 51 | 0.9020 | 0.0000 | 0.0000 | 0.3715 |
+| retail_stocking | 38 | 0.9474 | 0.0526 | 0.0000 | 0.2222 |
+
+## Train-Seen Split
+
+| group | samples | parsed_prediction_rate | action_exact_rate | next_action_exact_rate |
+| --- | --- | --- | --- | --- |
+| unseen_in_train | 317 | 0.4890 | 0.0000 | 0.0095 |
+| seen_in_train | 131 | 0.5649 | 0.0305 | 0.0229 |
+
+## Required-Modality State
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| rrd_missing_only_required_modalities_present | 448 | 0.5112 | 0.0089 | 0.1370 |
+
+## Object Categories
+
+| group | samples | object_precision | object_recall | object_f1 |
+| --- | --- | --- | --- | --- |
+| tool_stationery | 138 | 0.2642 | 0.0852 | 0.1288 |
+| craft_small_object | 106 | 0.3204 | 0.1065 | 0.1598 |
+| paper_cardboard | 261 | 0.2971 | 0.1067 | 0.1570 |
+| no_object_label | 2 | 0.0000 | 0.0000 | 0.0000 |
+| furniture_room | 96 | 0.1950 | 0.0978 | 0.1303 |
+| phone_device | 162 | 0.5245 | 0.1256 | 0.2027 |
+| food_kitchen | 56 | 0.2424 | 0.0711 | 0.1100 |
+| other_object | 135 | 0.1185 | 0.0595 | 0.0792 |
+
+## Interpretation
+
+The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality.
+
+Generated files:
+
+- `error_analysis_summary.json`
+- `episode_error_analysis.csv`
+- `action_family_error_analysis.csv`
+- `train_seen_error_analysis.csv`
+- `missing_modality_error_analysis.csv`
+- `object_category_error_analysis.csv`

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/action_family_error_analysis.csv

@@ -0,0 +1,9 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+small_object_sorting,87,0.3218390804597701,0.0,0.0,0.25287356321839083,0.0,0.12643678160919541,0.32786885245901637,0.08097165991902834,0.12987012987012986
+paper_cardboard_craft,142,0.323943661971831,0.014084507042253521,0.0,0.2746478873239437,0.014084507042253521,0.2746478873239437,0.19696969696969696,0.053388090349075976,0.0840064620355412
+other,94,0.5106382978723404,0.0,0.0,0.30851063829787234,0.0,0.2978723404255319,0.12179487179487179,0.07142857142857142,0.09004739336492891
+food_kitchen,5,0.6,0.0,0.0,0.2,0.0,0.0,0.0,0.0,0.0
+locomotion,23,0.6956521739130435,0.0,0.0,0.13043478260869565,0.0,0.21739130434782608,0.047619047619047616,0.0273972602739726,0.034782608695652174
+cleaning,8,0.75,0.0,0.0,0.375,0.0,0.375,0.0,0.0,0.0
+phone_use,51,0.9019607843137255,0.0,0.0,0.803921568627451,0.0,0.6078431372549019,0.573170731707317,0.27485380116959063,0.3715415019762846
+retail_stocking,38,0.9473684210526315,0.05263157894736842,0.0,0.7105263157894737,0.10526315789473684,0.7105263157894737,0.2289156626506024,0.2159090909090909,0.2222222222222222

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/episode_error_analysis.csv

@@ -0,0 +1,15 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,32,0.09375,0.0,0.0,0.09375,0.0,0.09375,0.2,0.017699115044247787,0.03252032520325203
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,32,0.25,0.0,0.0,0.09375,0.0,0.09375,0.0,0.0,0.0
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,32,0.28125,0.0,0.0,0.25,0.0,0.28125,0.39285714285714285,0.08527131782945736,0.14012738853503187
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,32,0.28125,0.03125,0.0,0.25,0.0,0.03125,0.38461538461538464,0.08849557522123894,0.14388489208633096
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4,32,0.3125,0.0,0.0,0.3125,0.0,0.125,0.7058823529411765,0.12903225806451613,0.21818181818181817
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,32,0.4375,0.0,0.0,0.25,0.03125,0.40625,0.0,0.0,0.0
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,32,0.46875,0.0,0.0,0.21875,0.0,0.21875,0.078125,0.045871559633027525,0.057803468208092484
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,32,0.53125,0.0,0.0,0.5,0.0,0.375,0.8235294117647058,0.09655172413793103,0.17283950617283952
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,32,0.65625,0.0,0.0,0.59375,0.03125,0.625,0.20987654320987653,0.1717171717171717,0.18888888888888888
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,32,0.65625,0.03125,0.0,0.53125,0.0,0.25,0.15730337078651685,0.12962962962962962,0.14213197969543145
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,32,0.6875,0.0,0.0,0.5625,0.0,0.40625,0.84,0.21649484536082475,0.3442622950819672
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,32,0.75,0.0,0.0,0.28125,0.0,0.28125,0.018518518518518517,0.0125,0.01492537313432836
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,32,0.84375,0.03125,0.0,0.375,0.0625,0.53125,0.08196721311475409,0.06666666666666667,0.07352941176470587
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,32,0.90625,0.03125,0.0,0.84375,0.0625,0.78125,0.3620689655172414,0.2916666666666667,0.3230769230769231

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/error_analysis_summary.json

@@ -0,0 +1,527 @@
+{
+  "status": "pass",
+  "model_label": "Cosmos3-Super Reasoner base-weight",
+  "source_package": "xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607",
+  "source_prediction_rows": 448,
+  "metrics_json_validity_rate": 0.5111607142857143,
+  "computed": {
+    "group": "overall",
+    "samples": 448,
+    "parsed_prediction_rate": 0.5111607142857143,
+    "action_exact_rate": 0.008928571428571428,
+    "subtask_exact_rate": 0.0,
+    "transition_exact_rate": 0.36830357142857145,
+    "next_action_exact_rate": 0.013392857142857142,
+    "contact_exact_rate": 0.32142857142857145,
+    "object_precision": 0.23130434782608697,
+    "object_recall": 0.09736456808199122,
+    "object_f1": 0.13704276146316333
+  },
+  "worst_episode_groups": [
+    {
+      "group": "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 0.09375,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.09375,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.09375,
+      "object_precision": 0.2,
+      "object_recall": 0.017699115044247787,
+      "object_f1": 0.03252032520325203
+    },
+    {
+      "group": "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6",
+      "samples": 32,
+      "parsed_prediction_rate": 0.25,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.09375,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.09375,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3",
+      "samples": 32,
+      "parsed_prediction_rate": 0.28125,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.25,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.28125,
+      "object_precision": 0.39285714285714285,
+      "object_recall": 0.08527131782945736,
+      "object_f1": 0.14012738853503187
+    },
+    {
+      "group": "b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 0.28125,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.25,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.03125,
+      "object_precision": 0.38461538461538464,
+      "object_recall": 0.08849557522123894,
+      "object_f1": 0.14388489208633096
+    },
+    {
+      "group": "877779cd-25f3-4293-a3c4-39067dd9558c__ep4",
+      "samples": 32,
+      "parsed_prediction_rate": 0.3125,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.3125,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.125,
+      "object_precision": 0.7058823529411765,
+      "object_recall": 0.12903225806451613,
+      "object_f1": 0.21818181818181817
+    },
+    {
+      "group": "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 0.4375,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.25,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.40625,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 0.46875,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.21875,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.21875,
+      "object_precision": 0.078125,
+      "object_recall": 0.045871559633027525,
+      "object_f1": 0.057803468208092484
+    },
+    {
+      "group": "ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 0.53125,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.5,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.375,
+      "object_precision": 0.8235294117647058,
+      "object_recall": 0.09655172413793103,
+      "object_f1": 0.17283950617283952
+    }
+  ],
+  "action_family_groups": [
+    {
+      "group": "small_object_sorting",
+      "samples": 87,
+      "parsed_prediction_rate": 0.3218390804597701,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.25287356321839083,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.12643678160919541,
+      "object_precision": 0.32786885245901637,
+      "object_recall": 0.08097165991902834,
+      "object_f1": 0.12987012987012986
+    },
+    {
+      "group": "paper_cardboard_craft",
+      "samples": 142,
+      "parsed_prediction_rate": 0.323943661971831,
+      "action_exact_rate": 0.014084507042253521,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.2746478873239437,
+      "next_action_exact_rate": 0.014084507042253521,
+      "contact_exact_rate": 0.2746478873239437,
+      "object_precision": 0.19696969696969696,
+      "object_recall": 0.053388090349075976,
+      "object_f1": 0.0840064620355412
+    },
+    {
+      "group": "other",
+      "samples": 94,
+      "parsed_prediction_rate": 0.5106382978723404,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.30851063829787234,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.2978723404255319,
+      "object_precision": 0.12179487179487179,
+      "object_recall": 0.07142857142857142,
+      "object_f1": 0.09004739336492891
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 5,
+      "parsed_prediction_rate": 0.6,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.2,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.0,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "locomotion",
+      "samples": 23,
+      "parsed_prediction_rate": 0.6956521739130435,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.13043478260869565,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.21739130434782608,
+      "object_precision": 0.047619047619047616,
+      "object_recall": 0.0273972602739726,
+      "object_f1": 0.034782608695652174
+    },
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 0.75,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.375,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.375,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "phone_use",
+      "samples": 51,
+      "parsed_prediction_rate": 0.9019607843137255,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.803921568627451,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.6078431372549019,
+      "object_precision": 0.573170731707317,
+      "object_recall": 0.27485380116959063,
+      "object_f1": 0.3715415019762846
+    },
+    {
+      "group": "retail_stocking",
+      "samples": 38,
+      "parsed_prediction_rate": 0.9473684210526315,
+      "action_exact_rate": 0.05263157894736842,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.7105263157894737,
+      "next_action_exact_rate": 0.10526315789473684,
+      "contact_exact_rate": 0.7105263157894737,
+      "object_precision": 0.2289156626506024,
+      "object_recall": 0.2159090909090909,
+      "object_f1": 0.2222222222222222
+    }
+  ],
+  "train_seen_groups": [
+    {
+      "group": "unseen_in_train",
+      "samples": 317,
+      "parsed_prediction_rate": 0.4889589905362776,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.3280757097791798,
+      "next_action_exact_rate": 0.00946372239747634,
+      "contact_exact_rate": 0.2902208201892745,
+      "object_precision": 0.18686868686868688,
+      "object_recall": 0.07905982905982906,
+      "object_f1": 0.1111111111111111
+    },
+    {
+      "group": "seen_in_train",
+      "samples": 131,
+      "parsed_prediction_rate": 0.5648854961832062,
+      "action_exact_rate": 0.030534351145038167,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.46564885496183206,
+      "next_action_exact_rate": 0.022900763358778626,
+      "contact_exact_rate": 0.3969465648854962,
+      "object_precision": 0.329608938547486,
+      "object_recall": 0.1372093023255814,
+      "object_f1": 0.19376026272577998
+    }
+  ],
+  "missing_modality_groups": [
+    {
+      "group": "rrd_missing_only_required_modalities_present",
+      "samples": 448,
+      "parsed_prediction_rate": 0.5111607142857143,
+      "action_exact_rate": 0.008928571428571428,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.36830357142857145,
+      "next_action_exact_rate": 0.013392857142857142,
+      "contact_exact_rate": 0.32142857142857145,
+      "object_precision": 0.23130434782608697,
+      "object_recall": 0.09736456808199122,
+      "object_f1": 0.13704276146316333
+    }
+  ],
+  "object_category_groups": [
+    {
+      "group": "tool_stationery",
+      "samples": 138,
+      "parsed_prediction_rate": 0.32608695652173914,
+      "action_exact_rate": 0.007246376811594203,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.2898550724637681,
+      "next_action_exact_rate": 0.007246376811594203,
+      "contact_exact_rate": 0.2536231884057971,
+      "object_precision": 0.2641509433962264,
+      "object_recall": 0.08519269776876268,
+      "object_f1": 0.12883435582822086
+    },
+    {
+      "group": "craft_small_object",
+      "samples": 106,
+      "parsed_prediction_rate": 0.3867924528301887,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.3113207547169811,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.1509433962264151,
+      "object_precision": 0.32038834951456313,
+      "object_recall": 0.1064516129032258,
+      "object_f1": 0.15980629539951574
+    },
+    {
+      "group": "paper_cardboard",
+      "samples": 261,
+      "parsed_prediction_rate": 0.4904214559386973,
+      "action_exact_rate": 0.011494252873563218,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.40229885057471265,
+      "next_action_exact_rate": 0.01532567049808429,
+      "contact_exact_rate": 0.36398467432950193,
+      "object_precision": 0.2971246006389776,
+      "object_recall": 0.10665137614678899,
+      "object_f1": 0.15696202531645567
+    },
+    {
+      "group": "no_object_label",
+      "samples": 2,
+      "parsed_prediction_rate": 0.5,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.5,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.5,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "furniture_room",
+      "samples": 96,
+      "parsed_prediction_rate": 0.5,
+      "action_exact_rate": 0.010416666666666666,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.2604166666666667,
+      "next_action_exact_rate": 0.010416666666666666,
+      "contact_exact_rate": 0.23958333333333334,
+      "object_precision": 0.1949685534591195,
+      "object_recall": 0.09779179810725552,
+      "object_f1": 0.13025210084033614
+    },
+    {
+      "group": "phone_device",
+      "samples": 162,
+      "parsed_prediction_rate": 0.5123456790123457,
+      "action_exact_rate": 0.006172839506172839,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.4567901234567901,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.3333333333333333,
+      "object_precision": 0.5244755244755245,
+      "object_recall": 0.12562814070351758,
+      "object_f1": 0.2027027027027027
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 56,
+      "parsed_prediction_rate": 0.5178571428571429,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.44642857142857145,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.35714285714285715,
+      "object_precision": 0.24242424242424243,
+      "object_recall": 0.07111111111111111,
+      "object_f1": 0.1099656357388316
+    },
+    {
+      "group": "other_object",
+      "samples": 135,
+      "parsed_prediction_rate": 0.5703703703703704,
+      "action_exact_rate": 0.007407407407407408,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.34814814814814815,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.3111111111111111,
+      "object_precision": 0.11848341232227488,
+      "object_recall": 0.05952380952380952,
+      "object_f1": 0.07923930269413627
+    },
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 0.625,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.375,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "retail_container",
+      "samples": 101,
+      "parsed_prediction_rate": 0.6732673267326733,
+      "action_exact_rate": 0.0297029702970297,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.5247524752475248,
+      "next_action_exact_rate": 0.039603960396039604,
+      "contact_exact_rate": 0.5346534653465347,
+      "object_precision": 0.2275449101796407,
+      "object_recall": 0.1148036253776435,
+      "object_f1": 0.15261044176706828
+    }
+  ],
+  "invalid_json_examples": [
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:90",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Move towards the stove",
+      "raw_prediction_prefix": "{\"action\": \"Browsing smartphone content\", \"subtask\": \"Browsing smartphone content\", \"objects\": [\"smartphone\", \" Mahjong tiles\", \" Mahjong tile stack\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\"], \"co"
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:100",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Open stove pot lid",
+      "raw_prediction_prefix": "{\"action\": \"Browsing smartphone content\", \"subtask\": \"Browsing smartphone content\", \"objects\": [\"smartphone\", \"hand\", \"kitchen counter\", \"sink\", \"window\", \"cabinet\", \"floor\", \"wall\", \"ceiling\", \"light fixture\", \"sink faucet\", \"kitchen sink\""
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:120",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Walking towards door",
+      "raw_prediction_prefix": "{\"action\": \"Browsing smartphone content\", \"subtask\": \"Browsing smartphone content\", \"objects\": [\"smartphone\", \"hand\", \"table\", \"chair\", \"floor\", \"wall\", \"window\", \"door\", \"light\", \"plant\", \"carpet\", \"railing\", \"balcony\", \"sky\", \"clouds\", \"t"
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:126",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Closing the door",
+      "raw_prediction_prefix": "{\"action\":\"Approach packing area\", \"subtask\":unknown, \"objects\":[\"packing area\", \"items\", \"containers\"], \"contact\":unknown, \"transition\":unknown, \"next_action\":unknown, \"evidence_window\":{\"start_frame\":0, \"end_frame\":1}}"
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:140",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Wipe kitchen counter",
+      "raw_prediction_prefix": "{\"action\": \"Browsing smartphone content\", \"subtask\": \"Browsing smartphone content\", \"objects\": [\"smartphone\", \" Mahjong tiles\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\", \" Mahjong tile\"], \"contact\""
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:166",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Place cloth on floor",
+      "raw_prediction_prefix": "{\"action\":\"unknown\",\"subtask\":\"unknown\",\"objects\":[]/\"contact\":\"unknown\",\"transition\":\"unknown\",\"next_action\":\"unknown\",\"evidence_window\":{\"start_frame\":0,\"end_frame\":0}}"
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:205",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Grasping cleaning cloth",
+      "raw_prediction_prefix": "{\"action\": \"Wipe kitchen counter\", \"subtask\": \"Cleaning kitchen surfaces\", \"objects\": [\"counter\", \"cloth\", \"sink\", \"window\", \"kitchen items\", \"red bowl\", \"green vegetables\", \"metal utensils\", \"blue container\", \"white bottle\", \"yellow sponge"
+    },
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:220",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Lift pot lid",
+      "raw_prediction_prefix": "{\"action\": \"Approach the stove\", \"subtask\": \"Approaching the stove\", \"objects\": [\"stove\", \"pot\", \"kitchen counter\", \"utensils\", \"person\", \"window\", \"sink\", \"cabinet\", \"floor\", \"wall\", \"ceiling\", \"light\", \"fan\", \"plant\", \"bottle\", \"can\", \"bo"
+    },
+    {
+      "id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:12",
+      "episode_id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "true_action": "Pick up utility knife",
+      "raw_prediction_prefix": "{\"action\": \"Adjust cardboard\", \"subtask\": \"Adjusting cardboard layout\", \"objects\": [\"cardboard\", \"ruler\", \"marker\", \"scissors\", \"work table\", \"measuring tape\", \"glue stick\", \"pencil\"], \"contact\": \"yes\", \"transition\": \"no\", \"next_action\": \"M"
+    },
+    {
+      "id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:13",
+      "episode_id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "true_action": "Pick up utility knife",
+      "raw_prediction_prefix": "{\"action\": \"Use phone\", \"subtask\": \"Use phone to check instructions\", \"objects\": [\"smartphone\", \"cardboard piece\", \"marker\", \"ruler\", \"scissors\", \"pen\", \"table\", \"workstation\", \"workspace\", \"craft materials\", \"paper strips\", \"quilling paper"
+    },
+    {
+      "id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:14",
+      "episode_id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "true_action": "Pick up utility knife",
+      "raw_prediction_prefix": "{\"action\": \"Approach packing area\", \"subtask\": \"Approaching packing area\", \"objects\": [\"cardboard piece\", \"scissors\", \"marker\", \"ruler\", \"work table\", \"packing area\", \"craft supplies\", \"tools\"], \"contact\": \"yes\", \"transition\": \"yes\", \"next_"
+    },
+    {
+      "id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1:qa:17",
+      "episode_id": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "true_action": "Pick up utility knife",
+      "raw_prediction_prefix": "{\"action\": \"Adjust cardboard\", \"subtask\": \"Adjusting cardboard layout\", \"objects\": [\"cardboard piece\", \"ruler\", \"marker\", \"scissors\", \"work table\", \"measuring tape\", \"cutting mat\", \"glue stick\"], \"contact\": \"yes\", \"transition\": \"no\", \"next_"
+    }
+  ],
+  "object_overgeneration_examples": [],
+  "modality_missing_by_episode": {
+    "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1": [
+      "visualization.rrd"
+    ],
+    "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1": [
+      "visualization.rrd"
+    ],
+    "33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1": [
+      "visualization.rrd"
+    ],
+    "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2": [
+      "visualization.rrd"
+    ],
+    "34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5": [
+      "visualization.rrd"
+    ],
+    "b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2": [
+      "visualization.rrd"
+    ],
+    "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2": [
+      "visualization.rrd"
+    ],
+    "4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1": [
+      "visualization.rrd"
+    ],
+    "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6": [
+      "visualization.rrd"
+    ],
+    "b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1": [
+      "visualization.rrd"
+    ],
+    "877779cd-25f3-4293-a3c4-39067dd9558c__ep4": [
+      "visualization.rrd"
+    ],
+    "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1": [
+      "visualization.rrd"
+    ],
+    "ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2": [
+      "visualization.rrd"
+    ],
+    "b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3": [
+      "visualization.rrd"
+    ]
+  },
+  "interpretation": "The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality."
+}

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/missing_modality_error_analysis.csv

@@ -0,0 +1,2 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+rrd_missing_only_required_modalities_present,448,0.5111607142857143,0.008928571428571428,0.0,0.36830357142857145,0.013392857142857142,0.32142857142857145,0.23130434782608697,0.09736456808199122,0.13704276146316333

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/object_category_error_analysis.csv

@@ -0,0 +1,11 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+tool_stationery,138,0.32608695652173914,0.007246376811594203,0.0,0.2898550724637681,0.007246376811594203,0.2536231884057971,0.2641509433962264,0.08519269776876268,0.12883435582822086
+craft_small_object,106,0.3867924528301887,0.0,0.0,0.3113207547169811,0.0,0.1509433962264151,0.32038834951456313,0.1064516129032258,0.15980629539951574
+paper_cardboard,261,0.4904214559386973,0.011494252873563218,0.0,0.40229885057471265,0.01532567049808429,0.36398467432950193,0.2971246006389776,0.10665137614678899,0.15696202531645567
+no_object_label,2,0.5,0.0,0.0,0.5,0.0,0.5,0.0,0.0,0.0
+furniture_room,96,0.5,0.010416666666666666,0.0,0.2604166666666667,0.010416666666666666,0.23958333333333334,0.1949685534591195,0.09779179810725552,0.13025210084033614
+phone_device,162,0.5123456790123457,0.006172839506172839,0.0,0.4567901234567901,0.0,0.3333333333333333,0.5244755244755245,0.12562814070351758,0.2027027027027027
+food_kitchen,56,0.5178571428571429,0.0,0.0,0.44642857142857145,0.0,0.35714285714285715,0.24242424242424243,0.07111111111111111,0.1099656357388316
+other_object,135,0.5703703703703704,0.007407407407407408,0.0,0.34814814814814815,0.0,0.3111111111111111,0.11848341232227488,0.05952380952380952,0.07923930269413627
+cleaning,8,0.625,0.0,0.0,0.0,0.0,0.375,0.0,0.0,0.0
+retail_container,101,0.6732673267326733,0.0297029702970297,0.0,0.5247524752475248,0.039603960396039604,0.5346534653465347,0.2275449101796407,0.1148036253776435,0.15261044176706828

results/omni_finetune/verified_public/xperience10m_cosmos3_super_reasoner_128ep_test_full_20260607/analysis/train_seen_error_analysis.csv

@@ -0,0 +1,3 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+unseen_in_train,317,0.4889589905362776,0.0,0.0,0.3280757097791798,0.00946372239747634,0.2902208201892745,0.18686868686868688,0.07905982905982906,0.1111111111111111
+seen_in_train,131,0.5648854961832062,0.030534351145038167,0.0,0.46564885496183206,0.022900763358778626,0.3969465648854962,0.329608938547486,0.1372093023255814,0.19376026272577998

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/ERROR_ANALYSIS.md

@@ -0,0 +1,78 @@
+# Qwen3-Omni LoRA v2 strict-json Held-Out Error Analysis
+
+This report is computed from the verified public package predictions. It contains only derived metrics and sanitized examples.
+
+## Overall
+
+- Prediction rows: `448`
+- JSON validity from `metrics.json`: `0.9978`
+- Parsed prediction rate from public rows: `0.9978`
+- Action exact rate: `0.0290`
+- Subtask exact rate: `0.0022`
+- Contact exact rate: `0.7188`
+- Object F1: `0.3016`
+
+## Weakest Episode Groups
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| 8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1 | 32 | 0.9688 | 0.0312 | 0.1677 |
+| 9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2 | 32 | 1.0000 | 0.0000 | 0.3745 |
+| b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2 | 32 | 1.0000 | 0.0000 | 0.4016 |
+| 5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6 | 32 | 1.0000 | 0.0000 | 0.0286 |
+| 877779cd-25f3-4293-a3c4-39067dd9558c__ep4 | 32 | 1.0000 | 0.0000 | 0.3587 |
+| 1796b943-caad-43c6-b9bd-80b8d601f37d__ep1 | 32 | 1.0000 | 0.0000 | 0.1244 |
+| a1012a57-385e-45a9-8a59-694a26fe92a5__ep1 | 32 | 1.0000 | 0.0312 | 0.5714 |
+| ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2 | 32 | 1.0000 | 0.0312 | 0.1205 |
+
+## Action Families
+
+| group | samples | parsed_prediction_rate | action_exact_rate | subtask_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- | --- |
+| cleaning | 8 | 0.8750 | 0.0000 | 0.0000 | 0.0455 |
+| locomotion | 23 | 1.0000 | 0.0000 | 0.0000 | 0.1250 |
+| small_object_sorting | 87 | 1.0000 | 0.0000 | 0.0000 | 0.2707 |
+| other | 94 | 1.0000 | 0.0000 | 0.0000 | 0.2803 |
+| phone_use | 51 | 1.0000 | 0.0196 | 0.0000 | 0.4132 |
+| paper_cardboard_craft | 142 | 1.0000 | 0.0493 | 0.0070 | 0.3580 |
+| retail_stocking | 38 | 1.0000 | 0.1053 | 0.0000 | 0.1582 |
+| food_kitchen | 5 | 1.0000 | 0.2000 | 0.0000 | 0.1538 |
+
+## Train-Seen Split
+
+| group | samples | parsed_prediction_rate | action_exact_rate | next_action_exact_rate |
+| --- | --- | --- | --- | --- |
+| unseen_in_train | 317 | 0.9968 | 0.0126 | 0.0126 |
+| seen_in_train | 131 | 1.0000 | 0.0687 | 0.0687 |
+
+## Required-Modality State
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| rrd_missing_only_required_modalities_present | 448 | 0.9978 | 0.0290 | 0.3016 |
+
+## Object Categories
+
+| group | samples | object_precision | object_recall | object_f1 |
+| --- | --- | --- | --- | --- |
+| furniture_room | 96 | 0.3356 | 0.3155 | 0.3252 |
+| other_object | 135 | 0.2000 | 0.2048 | 0.2024 |
+| no_object_label | 2 | 0.0000 | 0.0000 | 0.0000 |
+| cleaning | 8 | 0.0455 | 0.0476 | 0.0465 |
+| craft_small_object | 106 | 0.2508 | 0.2645 | 0.2575 |
+| tool_stationery | 138 | 0.4413 | 0.4422 | 0.4417 |
+| food_kitchen | 56 | 0.3202 | 0.2889 | 0.3037 |
+| phone_device | 162 | 0.3915 | 0.3719 | 0.3814 |
+
+## Interpretation
+
+The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality.
+
+Generated files:
+
+- `error_analysis_summary.json`
+- `episode_error_analysis.csv`
+- `action_family_error_analysis.csv`
+- `train_seen_error_analysis.csv`
+- `missing_modality_error_analysis.csv`
+- `object_category_error_analysis.csv`

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/action_family_error_analysis.csv

@@ -0,0 +1,9 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+cleaning,8,0.875,0.0,0.0,0.75,0.0,0.5,0.041666666666666664,0.05,0.04545454545454545
+locomotion,23,1.0,0.0,0.0,1.0,0.0,0.4782608695652174,0.1267605633802817,0.1232876712328767,0.125
+small_object_sorting,87,1.0,0.0,0.0,0.9885057471264368,0.0,0.6551724137931034,0.2701612903225806,0.27125506072874495,0.27070707070707073
+other,94,1.0,0.0,0.0,0.9574468085106383,0.0,0.6595744680851063,0.25705329153605017,0.3082706766917293,0.28034188034188035
+phone_use,51,1.0,0.0196078431372549,0.0,0.9411764705882353,0.0196078431372549,0.6470588235294118,0.4233128834355828,0.40350877192982454,0.4131736526946108
+paper_cardboard_craft,142,1.0,0.04929577464788732,0.007042253521126761,0.9859154929577465,0.04929577464788732,0.8802816901408451,0.3489278752436647,0.3675564681724846,0.35800000000000004
+retail_stocking,38,1.0,0.10526315789473684,0.0,0.9736842105263158,0.10526315789473684,0.7368421052631579,0.15730337078651685,0.1590909090909091,0.1581920903954802
+food_kitchen,5,1.0,0.2,0.0,1.0,0.2,0.4,0.16666666666666666,0.14285714285714285,0.15384615384615383

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/episode_error_analysis.csv

@@ -0,0 +1,15 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,32,0.96875,0.03125,0.0,0.9375,0.03125,0.59375,0.16091954022988506,0.175,0.16766467065868262
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,32,1.0,0.0,0.0,1.0,0.0,1.0,0.34057971014492755,0.415929203539823,0.37450199203187257
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,32,1.0,0.0,0.0,0.9375,0.0,0.40625,0.37404580152671757,0.4336283185840708,0.40163934426229503
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,32,1.0,0.0,0.0,1.0,0.0,1.0,0.029411764705882353,0.027777777777777776,0.02857142857142857
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4,32,1.0,0.0,0.0,1.0,0.0,0.40625,0.3626373626373626,0.3548387096774194,0.3586956521739131
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,32,1.0,0.0,0.0,1.0,0.0,0.5625,0.13,0.11926605504587157,0.1244019138755981
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,32,1.0,0.03125,0.0,1.0,0.03125,0.90625,0.6,0.5454545454545454,0.5714285714285713
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,32,1.0,0.03125,0.03125,1.0,0.03125,0.90625,0.09523809523809523,0.16393442622950818,0.12048192771084339
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,32,1.0,0.03125,0.0,0.96875,0.03125,0.84375,0.4253731343283582,0.3931034482758621,0.4086021505376344
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,32,1.0,0.03125,0.0,0.96875,0.03125,1.0,0.5607476635514018,0.46511627906976744,0.5084745762711863
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,32,1.0,0.0625,0.0,0.84375,0.0625,0.625,0.0449438202247191,0.05333333333333334,0.048780487804878044
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,32,1.0,0.0625,0.0,1.0,0.0625,0.90625,0.23333333333333334,0.19444444444444445,0.21212121212121213
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,32,1.0,0.0625,0.0,0.9375,0.0625,0.40625,0.32432432432432434,0.3711340206185567,0.34615384615384615
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,32,1.0,0.0625,0.0,1.0,0.0625,0.5,0.234375,0.2777777777777778,0.2542372881355932

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/error_analysis_summary.json

@@ -0,0 +1,461 @@
+{
+  "status": "pass",
+  "model_label": "Qwen3-Omni LoRA v2 strict-json",
+  "source_package": "xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full",
+  "source_prediction_rows": 448,
+  "metrics_json_validity_rate": 0.9977678571428571,
+  "computed": {
+    "group": "overall",
+    "samples": 448,
+    "parsed_prediction_rate": 0.9977678571428571,
+    "action_exact_rate": 0.029017857142857144,
+    "subtask_exact_rate": 0.002232142857142857,
+    "transition_exact_rate": 0.9709821428571429,
+    "next_action_exact_rate": 0.029017857142857144,
+    "contact_exact_rate": 0.71875,
+    "object_precision": 0.2939541348158443,
+    "object_recall": 0.3096632503660322,
+    "object_f1": 0.30160427807486634
+  },
+  "worst_episode_groups": [
+    {
+      "group": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 0.96875,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9375,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.59375,
+      "object_precision": 0.16091954022988506,
+      "object_recall": 0.175,
+      "object_f1": 0.16766467065868262
+    },
+    {
+      "group": "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.34057971014492755,
+      "object_recall": 0.415929203539823,
+      "object_f1": 0.37450199203187257
+    },
+    {
+      "group": "b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9375,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.40625,
+      "object_precision": 0.37404580152671757,
+      "object_recall": 0.4336283185840708,
+      "object_f1": 0.40163934426229503
+    },
+    {
+      "group": "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.029411764705882353,
+      "object_recall": 0.027777777777777776,
+      "object_f1": 0.02857142857142857
+    },
+    {
+      "group": "877779cd-25f3-4293-a3c4-39067dd9558c__ep4",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.40625,
+      "object_precision": 0.3626373626373626,
+      "object_recall": 0.3548387096774194,
+      "object_f1": 0.3586956521739131
+    },
+    {
+      "group": "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.5625,
+      "object_precision": 0.13,
+      "object_recall": 0.11926605504587157,
+      "object_f1": 0.1244019138755981
+    },
+    {
+      "group": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.90625,
+      "object_precision": 0.6,
+      "object_recall": 0.5454545454545454,
+      "object_f1": 0.5714285714285713
+    },
+    {
+      "group": "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.03125,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.90625,
+      "object_precision": 0.09523809523809523,
+      "object_recall": 0.16393442622950818,
+      "object_f1": 0.12048192771084339
+    }
+  ],
+  "action_family_groups": [
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 0.875,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.75,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.5,
+      "object_precision": 0.041666666666666664,
+      "object_recall": 0.05,
+      "object_f1": 0.04545454545454545
+    },
+    {
+      "group": "locomotion",
+      "samples": 23,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.4782608695652174,
+      "object_precision": 0.1267605633802817,
+      "object_recall": 0.1232876712328767,
+      "object_f1": 0.125
+    },
+    {
+      "group": "small_object_sorting",
+      "samples": 87,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9885057471264368,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.6551724137931034,
+      "object_precision": 0.2701612903225806,
+      "object_recall": 0.27125506072874495,
+      "object_f1": 0.27070707070707073
+    },
+    {
+      "group": "other",
+      "samples": 94,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9574468085106383,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.6595744680851063,
+      "object_precision": 0.25705329153605017,
+      "object_recall": 0.3082706766917293,
+      "object_f1": 0.28034188034188035
+    },
+    {
+      "group": "phone_use",
+      "samples": 51,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0196078431372549,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9411764705882353,
+      "next_action_exact_rate": 0.0196078431372549,
+      "contact_exact_rate": 0.6470588235294118,
+      "object_precision": 0.4233128834355828,
+      "object_recall": 0.40350877192982454,
+      "object_f1": 0.4131736526946108
+    },
+    {
+      "group": "paper_cardboard_craft",
+      "samples": 142,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.04929577464788732,
+      "subtask_exact_rate": 0.007042253521126761,
+      "transition_exact_rate": 0.9859154929577465,
+      "next_action_exact_rate": 0.04929577464788732,
+      "contact_exact_rate": 0.8802816901408451,
+      "object_precision": 0.3489278752436647,
+      "object_recall": 0.3675564681724846,
+      "object_f1": 0.35800000000000004
+    },
+    {
+      "group": "retail_stocking",
+      "samples": 38,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.10526315789473684,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9736842105263158,
+      "next_action_exact_rate": 0.10526315789473684,
+      "contact_exact_rate": 0.7368421052631579,
+      "object_precision": 0.15730337078651685,
+      "object_recall": 0.1590909090909091,
+      "object_f1": 0.1581920903954802
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 5,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.2,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.2,
+      "contact_exact_rate": 0.4,
+      "object_precision": 0.16666666666666666,
+      "object_recall": 0.14285714285714285,
+      "object_f1": 0.15384615384615383
+    }
+  ],
+  "train_seen_groups": [
+    {
+      "group": "unseen_in_train",
+      "samples": 317,
+      "parsed_prediction_rate": 0.9968454258675079,
+      "action_exact_rate": 0.012618296529968454,
+      "subtask_exact_rate": 0.0031545741324921135,
+      "transition_exact_rate": 0.9747634069400631,
+      "next_action_exact_rate": 0.012618296529968454,
+      "contact_exact_rate": 0.6908517350157729,
+      "object_precision": 0.25897187196896215,
+      "object_recall": 0.28525641025641024,
+      "object_f1": 0.2714794102694459
+    },
+    {
+      "group": "seen_in_train",
+      "samples": 131,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.06870229007633588,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9618320610687023,
+      "next_action_exact_rate": 0.06870229007633588,
+      "contact_exact_rate": 0.7862595419847328,
+      "object_precision": 0.38235294117647056,
+      "object_recall": 0.3627906976744186,
+      "object_f1": 0.3723150357995227
+    }
+  ],
+  "missing_modality_groups": [
+    {
+      "group": "rrd_missing_only_required_modalities_present",
+      "samples": 448,
+      "parsed_prediction_rate": 0.9977678571428571,
+      "action_exact_rate": 0.029017857142857144,
+      "subtask_exact_rate": 0.002232142857142857,
+      "transition_exact_rate": 0.9709821428571429,
+      "next_action_exact_rate": 0.029017857142857144,
+      "contact_exact_rate": 0.71875,
+      "object_precision": 0.2939541348158443,
+      "object_recall": 0.3096632503660322,
+      "object_f1": 0.30160427807486634
+    }
+  ],
+  "object_category_groups": [
+    {
+      "group": "furniture_room",
+      "samples": 96,
+      "parsed_prediction_rate": 0.9895833333333334,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9791666666666666,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.5625,
+      "object_precision": 0.33557046979865773,
+      "object_recall": 0.31545741324921134,
+      "object_f1": 0.3252032520325203
+    },
+    {
+      "group": "other_object",
+      "samples": 135,
+      "parsed_prediction_rate": 0.9925925925925926,
+      "action_exact_rate": 0.014814814814814815,
+      "subtask_exact_rate": 0.007407407407407408,
+      "transition_exact_rate": 0.9629629629629629,
+      "next_action_exact_rate": 0.014814814814814815,
+      "contact_exact_rate": 0.674074074074074,
+      "object_precision": 0.2,
+      "object_recall": 0.20476190476190476,
+      "object_f1": 0.2023529411764706
+    },
+    {
+      "group": "no_object_label",
+      "samples": 2,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.75,
+      "object_precision": 0.045454545454545456,
+      "object_recall": 0.047619047619047616,
+      "object_f1": 0.046511627906976744
+    },
+    {
+      "group": "craft_small_object",
+      "samples": 106,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.009433962264150943,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9905660377358491,
+      "next_action_exact_rate": 0.009433962264150943,
+      "contact_exact_rate": 0.6132075471698113,
+      "object_precision": 0.25076452599388377,
+      "object_recall": 0.2645161290322581,
+      "object_f1": 0.2574568288854003
+    },
+    {
+      "group": "tool_stationery",
+      "samples": 138,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.014492753623188406,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9782608695652174,
+      "next_action_exact_rate": 0.014492753623188406,
+      "contact_exact_rate": 0.8333333333333334,
+      "object_precision": 0.44129554655870445,
+      "object_recall": 0.4421906693711968,
+      "object_f1": 0.4417426545086119
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 56,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.017857142857142856,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9464285714285714,
+      "next_action_exact_rate": 0.017857142857142856,
+      "contact_exact_rate": 0.7857142857142857,
+      "object_precision": 0.32019704433497537,
+      "object_recall": 0.28888888888888886,
+      "object_f1": 0.3037383177570093
+    },
+    {
+      "group": "phone_device",
+      "samples": 162,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.018518518518518517,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9567901234567902,
+      "next_action_exact_rate": 0.018518518518518517,
+      "contact_exact_rate": 0.6296296296296297,
+      "object_precision": 0.3915343915343915,
+      "object_recall": 0.37185929648241206,
+      "object_f1": 0.3814432989690722
+    },
+    {
+      "group": "paper_cardboard",
+      "samples": 261,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.038314176245210725,
+      "subtask_exact_rate": 0.0038314176245210726,
+      "transition_exact_rate": 0.9770114942528736,
+      "next_action_exact_rate": 0.038314176245210725,
+      "contact_exact_rate": 0.7662835249042146,
+      "object_precision": 0.361863488624052,
+      "object_recall": 0.3830275229357798,
+      "object_f1": 0.37214484679665744
+    },
+    {
+      "group": "retail_container",
+      "samples": 101,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.06930693069306931,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9603960396039604,
+      "next_action_exact_rate": 0.06930693069306931,
+      "contact_exact_rate": 0.8118811881188119,
+      "object_precision": 0.2542955326460481,
+      "object_recall": 0.22356495468277945,
+      "object_f1": 0.23794212218649516
+    }
+  ],
+  "invalid_json_examples": [
+    {
+      "id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1:qa:200",
+      "episode_id": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "true_action": "Washing hands in sink",
+      "raw_prediction_prefix": "{\"action\":\"Close door\",\"contact\":\"yes\",\"evidence_window\":{\"end_frame\":4019,\"start_frame\":4000},\"next_action\":\"Close door\",\"objects\":[\"door\",\"door handle\",\"bathroom counter\",\"bathroom sink\",\"bathroom cabinet\",\"bathroom mirror\",\"bathroom ligh"
+    }
+  ],
+  "object_overgeneration_examples": [],
+  "modality_missing_by_episode": {
+    "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1": [
+      "visualization.rrd"
+    ],
+    "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1": [
+      "visualization.rrd"
+    ],
+    "33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1": [
+      "visualization.rrd"
+    ],
+    "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2": [
+      "visualization.rrd"
+    ],
+    "34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5": [
+      "visualization.rrd"
+    ],
+    "b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2": [
+      "visualization.rrd"
+    ],
+    "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2": [
+      "visualization.rrd"
+    ],
+    "4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1": [
+      "visualization.rrd"
+    ],
+    "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6": [
+      "visualization.rrd"
+    ],
+    "b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1": [
+      "visualization.rrd"
+    ],
+    "877779cd-25f3-4293-a3c4-39067dd9558c__ep4": [
+      "visualization.rrd"
+    ],
+    "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1": [
+      "visualization.rrd"
+    ],
+    "ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2": [
+      "visualization.rrd"
+    ],
+    "b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3": [
+      "visualization.rrd"
+    ]
+  },
+  "interpretation": "The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality."
+}

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/missing_modality_error_analysis.csv

@@ -0,0 +1,2 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+rrd_missing_only_required_modalities_present,448,0.9977678571428571,0.029017857142857144,0.002232142857142857,0.9709821428571429,0.029017857142857144,0.71875,0.2939541348158443,0.3096632503660322,0.30160427807486634

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/object_category_error_analysis.csv

@@ -0,0 +1,11 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+furniture_room,96,0.9895833333333334,0.03125,0.0,0.9791666666666666,0.03125,0.5625,0.33557046979865773,0.31545741324921134,0.3252032520325203
+other_object,135,0.9925925925925926,0.014814814814814815,0.007407407407407408,0.9629629629629629,0.014814814814814815,0.674074074074074,0.2,0.20476190476190476,0.2023529411764706
+no_object_label,2,1.0,0.0,0.0,1.0,0.0,1.0,0.0,0.0,0.0
+cleaning,8,1.0,0.0,0.0,1.0,0.0,0.75,0.045454545454545456,0.047619047619047616,0.046511627906976744
+craft_small_object,106,1.0,0.009433962264150943,0.0,0.9905660377358491,0.009433962264150943,0.6132075471698113,0.25076452599388377,0.2645161290322581,0.2574568288854003
+tool_stationery,138,1.0,0.014492753623188406,0.0,0.9782608695652174,0.014492753623188406,0.8333333333333334,0.44129554655870445,0.4421906693711968,0.4417426545086119
+food_kitchen,56,1.0,0.017857142857142856,0.0,0.9464285714285714,0.017857142857142856,0.7857142857142857,0.32019704433497537,0.28888888888888886,0.3037383177570093
+phone_device,162,1.0,0.018518518518518517,0.0,0.9567901234567902,0.018518518518518517,0.6296296296296297,0.3915343915343915,0.37185929648241206,0.3814432989690722
+paper_cardboard,261,1.0,0.038314176245210725,0.0038314176245210726,0.9770114942528736,0.038314176245210725,0.7662835249042146,0.361863488624052,0.3830275229357798,0.37214484679665744
+retail_container,101,1.0,0.06930693069306931,0.0,0.9603960396039604,0.06930693069306931,0.8118811881188119,0.2542955326460481,0.22356495468277945,0.23794212218649516

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora_eval_test_full/analysis/train_seen_error_analysis.csv

@@ -0,0 +1,3 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+unseen_in_train,317,0.9968454258675079,0.012618296529968454,0.0031545741324921135,0.9747634069400631,0.012618296529968454,0.6908517350157729,0.25897187196896215,0.28525641025641024,0.2714794102694459
+seen_in_train,131,1.0,0.06870229007633588,0.0,0.9618320610687023,0.06870229007633588,0.7862595419847328,0.38235294117647056,0.3627906976744186,0.3723150357995227

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/PUBLIC_RESULT_SUMMARY.md

@@ -0,0 +1,25 @@
+# Verified Omni Fine-Tuning Result
+
+- Backbone: `qwen3_omni_lora`
+- Dataset run: `xperience10m_qwen3_omni_128ep_96train_16val_16test_valmon_20260605`
+- Training run: `xperience10m_qwen3_omni_128ep_structured_json_v2_reuse_full8gpu_lora`
+- Evaluation run: `xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full`
+- Validation status: `verified`
+- Held-out eval split: `test`
+- Held-out episodes: `14`
+- Prediction rows: `448`
+
+## Primary Metrics
+
+- json_validity_rate: `1.0`
+- action_macro_f1: `0.0021983997167007384`
+- subtask_accuracy: `0.002232142857142857`
+- transition_accuracy: `0.9732142857142857`
+- next_action_accuracy: `0.03125`
+- contact_accuracy: `0.7209821428571429`
+- object_micro_f1: `0.30688228657389993`
+- held_out_episode_count: `14`
+
+Raw Xperience-10M files, base-model weights, adapter or checkpoint weights, full checkpoints, and large archives are not included.
+
+Use this package as the source for README, website, and Hugging Face updates.

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/ERROR_ANALYSIS.md

@@ -0,0 +1,78 @@
+# Qwen3-Omni strict-label v3 Held-Out Error Analysis
+
+This report is computed from public-safe predictions and an episode manifest. It contains only derived metrics and sanitized examples.
+
+## Overall
+
+- Prediction rows: `448`
+- JSON validity from `metrics.json`: `1.0000`
+- Parsed prediction rate from public rows: `1.0000`
+- Action exact rate: `0.0312`
+- Subtask exact rate: `0.0022`
+- Contact exact rate: `0.7210`
+- Object F1: `0.3069`
+
+## Weakest Episode Groups
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| a1012a57-385e-45a9-8a59-694a26fe92a5__ep1 | 32 | 1.0000 | 0.0000 | 0.5341 |
+| 9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2 | 32 | 1.0000 | 0.0000 | 0.3719 |
+| ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2 | 32 | 1.0000 | 0.0000 | 0.1039 |
+| 5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6 | 32 | 1.0000 | 0.0000 | 0.0000 |
+| 877779cd-25f3-4293-a3c4-39067dd9558c__ep4 | 32 | 1.0000 | 0.0000 | 0.3315 |
+| 1796b943-caad-43c6-b9bd-80b8d601f37d__ep1 | 32 | 1.0000 | 0.0000 | 0.1347 |
+| 8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1 | 32 | 1.0000 | 0.0312 | 0.1928 |
+| 33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1 | 32 | 1.0000 | 0.0312 | 0.0774 |
+
+## Action Families
+
+| group | samples | parsed_prediction_rate | action_exact_rate | subtask_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- | --- |
+| cleaning | 8 | 1.0000 | 0.0000 | 0.0000 | 0.0435 |
+| locomotion | 23 | 1.0000 | 0.0000 | 0.0000 | 0.1471 |
+| small_object_sorting | 87 | 1.0000 | 0.0000 | 0.0000 | 0.2716 |
+| other | 94 | 1.0000 | 0.0106 | 0.0000 | 0.2831 |
+| phone_use | 51 | 1.0000 | 0.0392 | 0.0000 | 0.4175 |
+| paper_cardboard_craft | 142 | 1.0000 | 0.0493 | 0.0070 | 0.3559 |
+| retail_stocking | 38 | 1.0000 | 0.0789 | 0.0000 | 0.2093 |
+| food_kitchen | 5 | 1.0000 | 0.2000 | 0.0000 | 0.3333 |
+
+## Train-Seen Split
+
+| group | samples | parsed_prediction_rate | action_exact_rate | next_action_exact_rate |
+| --- | --- | --- | --- | --- |
+| unseen_in_train | 317 | 1.0000 | 0.0095 | 0.0095 |
+| seen_in_train | 131 | 1.0000 | 0.0840 | 0.0840 |
+
+## Required-Modality State
+
+| group | samples | parsed_prediction_rate | action_exact_rate | object_f1 |
+| --- | --- | --- | --- | --- |
+| rrd_missing_only_required_modalities_present | 448 | 1.0000 | 0.0312 | 0.3069 |
+
+## Object Categories
+
+| group | samples | object_precision | object_recall | object_f1 |
+| --- | --- | --- | --- | --- |
+| no_object_label | 2 | 0.0000 | 0.0000 | 0.0000 |
+| cleaning | 8 | 0.0500 | 0.0476 | 0.0488 |
+| craft_small_object | 106 | 0.2604 | 0.2419 | 0.2508 |
+| furniture_room | 96 | 0.3047 | 0.2681 | 0.2852 |
+| tool_stationery | 138 | 0.4528 | 0.4280 | 0.4400 |
+| other_object | 135 | 0.2288 | 0.2119 | 0.2200 |
+| food_kitchen | 56 | 0.3875 | 0.2756 | 0.3221 |
+| phone_device | 162 | 0.4435 | 0.3618 | 0.3985 |
+
+## Interpretation
+
+The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality.
+
+Generated files:
+
+- `error_analysis_summary.json`
+- `episode_error_analysis.csv`
+- `action_family_error_analysis.csv`
+- `train_seen_error_analysis.csv`
+- `missing_modality_error_analysis.csv`
+- `object_category_error_analysis.csv`

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/action_family_error_analysis.csv

@@ -0,0 +1,9 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+cleaning,8,1.0,0.0,0.0,0.875,0.0,0.625,0.038461538461538464,0.05,0.043478260869565216
+locomotion,23,1.0,0.0,0.0,1.0,0.0,0.43478260869565216,0.15873015873015872,0.136986301369863,0.14705882352941177
+small_object_sorting,87,1.0,0.0,0.0,0.9885057471264368,0.0,0.6551724137931034,0.27615062761506276,0.26720647773279355,0.2716049382716049
+other,94,1.0,0.010638297872340425,0.0,0.9574468085106383,0.010638297872340425,0.6595744680851063,0.28044280442804426,0.2857142857142857,0.28305400372439476
+phone_use,51,1.0,0.0392156862745098,0.0,0.9411764705882353,0.0392156862745098,0.6470588235294118,0.49206349206349204,0.36257309941520466,0.41750841750841755
+paper_cardboard_craft,142,1.0,0.04929577464788732,0.007042253521126761,0.9859154929577465,0.04929577464788732,0.8802816901408451,0.36075949367088606,0.351129363449692,0.35587929240374605
+retail_stocking,38,1.0,0.07894736842105263,0.0,0.9736842105263158,0.07894736842105263,0.7631578947368421,0.21428571428571427,0.20454545454545456,0.20930232558139536
+food_kitchen,5,1.0,0.2,0.0,1.0,0.2,0.4,0.4,0.2857142857142857,0.3333333333333333

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/episode_error_analysis.csv

@@ -0,0 +1,15 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+a1012a57-385e-45a9-8a59-694a26fe92a5__ep1,32,1.0,0.0,0.0,1.0,0.0,0.90625,0.6103896103896104,0.47474747474747475,0.5340909090909091
+9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2,32,1.0,0.0,0.03125,1.0,0.0,1.0,0.3488372093023256,0.39823008849557523,0.37190082644628103
+ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2,32,1.0,0.0,0.0,1.0,0.0,0.90625,0.08602150537634409,0.13114754098360656,0.1038961038961039
+5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6,32,1.0,0.0,0.0,1.0,0.0,1.0,0.0,0.0,0.0
+877779cd-25f3-4293-a3c4-39067dd9558c__ep4,32,1.0,0.0,0.0,1.0,0.0,0.40625,0.3409090909090909,0.3225806451612903,0.3314917127071823
+1796b943-caad-43c6-b9bd-80b8d601f37d__ep1,32,1.0,0.0,0.0,1.0,0.0,0.5,0.15476190476190477,0.11926605504587157,0.13471502590673576
+8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1,32,1.0,0.03125,0.0,0.96875,0.03125,0.625,0.18604651162790697,0.2,0.19277108433734938
+33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1,32,1.0,0.03125,0.0,0.84375,0.03125,0.6875,0.075,0.08,0.07741935483870968
+b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3,32,1.0,0.03125,0.0,0.96875,0.03125,1.0,0.5229357798165137,0.4418604651162791,0.4789915966386554
+34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5,32,1.0,0.0625,0.0,1.0,0.0625,0.90625,0.3157894736842105,0.25,0.27906976744186046
+b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2,32,1.0,0.0625,0.0,0.9375,0.0625,0.40625,0.4140625,0.4690265486725664,0.43983402489626555
+b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1,32,1.0,0.0625,0.0,1.0,0.0625,0.5,0.23853211009174313,0.24074074074074073,0.23963133640552997
+ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2,32,1.0,0.0625,0.0,0.96875,0.0625,0.84375,0.5346534653465347,0.3724137931034483,0.43902439024390244
+4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1,32,1.0,0.09375,0.0,0.9375,0.09375,0.40625,0.3977272727272727,0.36082474226804123,0.37837837837837834

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/error_analysis_summary.json

@@ -0,0 +1,456 @@
+{
+  "status": "pass",
+  "model_label": "Qwen3-Omni strict-label v3",
+  "source_package": "xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full",
+  "source_eval_dir": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/eval",
+  "source_episode_manifest": "results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/dataset/episode_manifest.json",
+  "source_prediction_rows": 448,
+  "metrics_json_validity_rate": 1.0,
+  "computed": {
+    "group": "overall",
+    "samples": 448,
+    "parsed_prediction_rate": 1.0,
+    "action_exact_rate": 0.03125,
+    "subtask_exact_rate": 0.002232142857142857,
+    "transition_exact_rate": 0.9732142857142857,
+    "next_action_exact_rate": 0.03125,
+    "contact_exact_rate": 0.7209821428571429,
+    "object_precision": 0.31554524361948955,
+    "object_recall": 0.2986822840409956,
+    "object_f1": 0.30688228657389993
+  },
+  "worst_episode_groups": [
+    {
+      "group": "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.90625,
+      "object_precision": 0.6103896103896104,
+      "object_recall": 0.47474747474747475,
+      "object_f1": 0.5340909090909091
+    },
+    {
+      "group": "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.03125,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.3488372093023256,
+      "object_recall": 0.39823008849557523,
+      "object_f1": 0.37190082644628103
+    },
+    {
+      "group": "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.90625,
+      "object_precision": 0.08602150537634409,
+      "object_recall": 0.13114754098360656,
+      "object_f1": 0.1038961038961039
+    },
+    {
+      "group": "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "877779cd-25f3-4293-a3c4-39067dd9558c__ep4",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.40625,
+      "object_precision": 0.3409090909090909,
+      "object_recall": 0.3225806451612903,
+      "object_f1": 0.3314917127071823
+    },
+    {
+      "group": "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.5,
+      "object_precision": 0.15476190476190477,
+      "object_recall": 0.11926605504587157,
+      "object_f1": 0.13471502590673576
+    },
+    {
+      "group": "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.96875,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.625,
+      "object_precision": 0.18604651162790697,
+      "object_recall": 0.2,
+      "object_f1": 0.19277108433734938
+    },
+    {
+      "group": "33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1",
+      "samples": 32,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.84375,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.6875,
+      "object_precision": 0.075,
+      "object_recall": 0.08,
+      "object_f1": 0.07741935483870968
+    }
+  ],
+  "action_family_groups": [
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.875,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.625,
+      "object_precision": 0.038461538461538464,
+      "object_recall": 0.05,
+      "object_f1": 0.043478260869565216
+    },
+    {
+      "group": "locomotion",
+      "samples": 23,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.43478260869565216,
+      "object_precision": 0.15873015873015872,
+      "object_recall": 0.136986301369863,
+      "object_f1": 0.14705882352941177
+    },
+    {
+      "group": "small_object_sorting",
+      "samples": 87,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9885057471264368,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.6551724137931034,
+      "object_precision": 0.27615062761506276,
+      "object_recall": 0.26720647773279355,
+      "object_f1": 0.2716049382716049
+    },
+    {
+      "group": "other",
+      "samples": 94,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.010638297872340425,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9574468085106383,
+      "next_action_exact_rate": 0.010638297872340425,
+      "contact_exact_rate": 0.6595744680851063,
+      "object_precision": 0.28044280442804426,
+      "object_recall": 0.2857142857142857,
+      "object_f1": 0.28305400372439476
+    },
+    {
+      "group": "phone_use",
+      "samples": 51,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0392156862745098,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9411764705882353,
+      "next_action_exact_rate": 0.0392156862745098,
+      "contact_exact_rate": 0.6470588235294118,
+      "object_precision": 0.49206349206349204,
+      "object_recall": 0.36257309941520466,
+      "object_f1": 0.41750841750841755
+    },
+    {
+      "group": "paper_cardboard_craft",
+      "samples": 142,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.04929577464788732,
+      "subtask_exact_rate": 0.007042253521126761,
+      "transition_exact_rate": 0.9859154929577465,
+      "next_action_exact_rate": 0.04929577464788732,
+      "contact_exact_rate": 0.8802816901408451,
+      "object_precision": 0.36075949367088606,
+      "object_recall": 0.351129363449692,
+      "object_f1": 0.35587929240374605
+    },
+    {
+      "group": "retail_stocking",
+      "samples": 38,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.07894736842105263,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9736842105263158,
+      "next_action_exact_rate": 0.07894736842105263,
+      "contact_exact_rate": 0.7631578947368421,
+      "object_precision": 0.21428571428571427,
+      "object_recall": 0.20454545454545456,
+      "object_f1": 0.20930232558139536
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 5,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.2,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.2,
+      "contact_exact_rate": 0.4,
+      "object_precision": 0.4,
+      "object_recall": 0.2857142857142857,
+      "object_f1": 0.3333333333333333
+    }
+  ],
+  "train_seen_groups": [
+    {
+      "group": "unseen_in_train",
+      "samples": 317,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.00946372239747634,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9779179810725552,
+      "next_action_exact_rate": 0.00946372239747634,
+      "contact_exact_rate": 0.694006309148265,
+      "object_precision": 0.2781954887218045,
+      "object_recall": 0.2767094017094017,
+      "object_f1": 0.27745045527584367
+    },
+    {
+      "group": "seen_in_train",
+      "samples": 131,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.08396946564885496,
+      "subtask_exact_rate": 0.007633587786259542,
+      "transition_exact_rate": 0.9618320610687023,
+      "next_action_exact_rate": 0.08396946564885496,
+      "contact_exact_rate": 0.7862595419847328,
+      "object_precision": 0.4116022099447514,
+      "object_recall": 0.34651162790697676,
+      "object_f1": 0.3762626262626263
+    }
+  ],
+  "missing_modality_groups": [
+    {
+      "group": "rrd_missing_only_required_modalities_present",
+      "samples": 448,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03125,
+      "subtask_exact_rate": 0.002232142857142857,
+      "transition_exact_rate": 0.9732142857142857,
+      "next_action_exact_rate": 0.03125,
+      "contact_exact_rate": 0.7209821428571429,
+      "object_precision": 0.31554524361948955,
+      "object_recall": 0.2986822840409956,
+      "object_f1": 0.30688228657389993
+    }
+  ],
+  "object_category_groups": [
+    {
+      "group": "no_object_label",
+      "samples": 2,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 1.0,
+      "object_precision": 0.0,
+      "object_recall": 0.0,
+      "object_f1": 0.0
+    },
+    {
+      "group": "cleaning",
+      "samples": 8,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.0,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 1.0,
+      "next_action_exact_rate": 0.0,
+      "contact_exact_rate": 0.75,
+      "object_precision": 0.05,
+      "object_recall": 0.047619047619047616,
+      "object_f1": 0.04878048780487805
+    },
+    {
+      "group": "craft_small_object",
+      "samples": 106,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.018867924528301886,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9905660377358491,
+      "next_action_exact_rate": 0.018867924528301886,
+      "contact_exact_rate": 0.6132075471698113,
+      "object_precision": 0.2604166666666667,
+      "object_recall": 0.24193548387096775,
+      "object_f1": 0.25083612040133774
+    },
+    {
+      "group": "furniture_room",
+      "samples": 96,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.020833333333333332,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9895833333333334,
+      "next_action_exact_rate": 0.020833333333333332,
+      "contact_exact_rate": 0.5520833333333334,
+      "object_precision": 0.3046594982078853,
+      "object_recall": 0.26813880126182965,
+      "object_f1": 0.28523489932885904
+    },
+    {
+      "group": "tool_stationery",
+      "samples": 138,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.021739130434782608,
+      "subtask_exact_rate": 0.007246376811594203,
+      "transition_exact_rate": 0.9782608695652174,
+      "next_action_exact_rate": 0.021739130434782608,
+      "contact_exact_rate": 0.8333333333333334,
+      "object_precision": 0.45278969957081544,
+      "object_recall": 0.4279918864097363,
+      "object_f1": 0.4400417101147028
+    },
+    {
+      "group": "other_object",
+      "samples": 135,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.022222222222222223,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9703703703703703,
+      "next_action_exact_rate": 0.022222222222222223,
+      "contact_exact_rate": 0.674074074074074,
+      "object_precision": 0.22879177377892032,
+      "object_recall": 0.2119047619047619,
+      "object_f1": 0.2200247218788628
+    },
+    {
+      "group": "food_kitchen",
+      "samples": 56,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.03571428571428571,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9464285714285714,
+      "next_action_exact_rate": 0.03571428571428571,
+      "contact_exact_rate": 0.7857142857142857,
+      "object_precision": 0.3875,
+      "object_recall": 0.27555555555555555,
+      "object_f1": 0.3220779220779221
+    },
+    {
+      "group": "phone_device",
+      "samples": 162,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.043209876543209874,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9567901234567902,
+      "next_action_exact_rate": 0.043209876543209874,
+      "contact_exact_rate": 0.6296296296296297,
+      "object_precision": 0.44353182751540043,
+      "object_recall": 0.36180904522613067,
+      "object_f1": 0.3985239852398524
+    },
+    {
+      "group": "paper_cardboard",
+      "samples": 261,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.04597701149425287,
+      "subtask_exact_rate": 0.0038314176245210726,
+      "transition_exact_rate": 0.9770114942528736,
+      "next_action_exact_rate": 0.04597701149425287,
+      "contact_exact_rate": 0.7739463601532567,
+      "object_precision": 0.3888213851761847,
+      "object_recall": 0.3669724770642202,
+      "object_f1": 0.3775811209439528
+    },
+    {
+      "group": "retail_container",
+      "samples": 101,
+      "parsed_prediction_rate": 1.0,
+      "action_exact_rate": 0.06930693069306931,
+      "subtask_exact_rate": 0.0,
+      "transition_exact_rate": 0.9603960396039604,
+      "next_action_exact_rate": 0.06930693069306931,
+      "contact_exact_rate": 0.8217821782178217,
+      "object_precision": 0.30833333333333335,
+      "object_recall": 0.22356495468277945,
+      "object_f1": 0.25919439579684767
+    }
+  ],
+  "invalid_json_examples": [],
+  "object_overgeneration_examples": [],
+  "modality_missing_by_episode": {
+    "8a8e1b3c-607e-4ada-b3fd-fa639727e92c__ep1": [
+      "visualization.rrd"
+    ],
+    "a1012a57-385e-45a9-8a59-694a26fe92a5__ep1": [
+      "visualization.rrd"
+    ],
+    "33f7ae08-ac1d-4321-9cb9-eca79016b359__ep1": [
+      "visualization.rrd"
+    ],
+    "9c553886-83c5-4dc4-be5c-dcb269b3a771__ep2": [
+      "visualization.rrd"
+    ],
+    "34f07a04-eb37-45a3-95ec-189ed5f4a85b__ep5": [
+      "visualization.rrd"
+    ],
+    "b9dd769b-e31a-4fdb-945e-5a60db6487b0__ep2": [
+      "visualization.rrd"
+    ],
+    "ba045ed4-ef25-404d-b756-8dcbd45b18fa__ep2": [
+      "visualization.rrd"
+    ],
+    "4b02bb38-384a-438a-b5f9-6131d85c34b0__ep1": [
+      "visualization.rrd"
+    ],
+    "5399ef86-4df9-49bc-809f-8f4f92f9e659__ep6": [
+      "visualization.rrd"
+    ],
+    "b750fab3-7fbb-43a0-b451-c64c4d4a64da__ep1": [
+      "visualization.rrd"
+    ],
+    "877779cd-25f3-4293-a3c4-39067dd9558c__ep4": [
+      "visualization.rrd"
+    ],
+    "1796b943-caad-43c6-b9bd-80b8d601f37d__ep1": [
+      "visualization.rrd"
+    ],
+    "ba18b7c1-21ff-45da-8452-41acce7fc8de__ep2": [
+      "visualization.rrd"
+    ],
+    "b6579cb5-0a71-4ca6-8808-1e2700be05c7__ep3": [
+      "visualization.rrd"
+    ]
+  },
+  "interpretation": "The diagnostic pilot is dominated by invalid or weak structured outputs and exact-label failures. These tables identify where to tighten JSON constraints, action/subtask target formatting, object vocabularies, and missing-modality robustness before claiming stronger model quality."
+}

results/omni_finetune/verified_public/xperience10m_qwen3_omni_128ep_structured_json_v3_strict_label_prompt_reuse_lora_eval_test_full/analysis/missing_modality_error_analysis.csv

@@ -0,0 +1,2 @@
+group,samples,parsed_prediction_rate,action_exact_rate,subtask_exact_rate,transition_exact_rate,next_action_exact_rate,contact_exact_rate,object_precision,object_recall,object_f1
+rrd_missing_only_required_modalities_present,448,1.0,0.03125,0.002232142857142857,0.9732142857142857,0.03125,0.7209821428571429,0.31554524361948955,0.2986822840409956,0.30688228657389993