Publish dLLM PRM Gap release

.gitattributes

@@ -0,0 +1,35 @@
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,55 @@
+---
+license: mit
+base_model: GSAI-ML/LLaDA-8B-Base
+tags:
+  - dllm-prm-gap
+  - discrete-diffusion
+  - gsm8k
+  - process-reward-model
+  - llada-8b
+pipeline_tag: text-classification
+library_name: peft
+---
+
+<h1 style="text-align: left; font-size: 1.6em; margin-bottom: 0.75em;">
+  <span style="color:#E69F00; font-weight:bold;">dLLM</span><span style="color:#0072B2; font-weight:bold;"> PRM</span><span style="color:#009E73; font-weight:bold;"> Gap</span> · Bidirectional LLaDA PRM
+</h1>
+
+<div style="text-align: left; margin-bottom: 18px;">
+  <a href="https://github.com/dLLM-PRM-Gap/dLLM-PRM-Gap" target="_blank">💻 Code</a>
+  &nbsp; • &nbsp;
+  <a href="https://huggingface.co/collections/YanZhanPKU/dllm-prm-gap" target="_blank">🤗 Collection</a>
+  &nbsp; • &nbsp;
+  <a href="https://arxiv.org/abs/2609.35472" target="_blank">📄 Paper</a>
+</div>
+
+## News
+
+- **2026-09**: Accepted at **NeurIPS 2026**.
+
+Public adapter release for **dLLM PRM Gap**. bidirectional PRM for the cross-backbone LLaDA diagnostic
+
+> This repository contains only trainable adapter parameters and the reward head. It does not include the base model. The release configuration is recorded in `config.json`; the arXiv paper defines the release scope and citation.
+
+## Files
+
+- `adapter.safetensors` — compact adapter weights
+- `config.json` — public base-model id, architecture, and provenance
+
+## Load
+
+```python
+from huggingface_hub import snapshot_download
+from prm.checkpointing import load_diffusion_prm
+
+path = snapshot_download("YanZhanPKU/dLLM-PRM-Gap-bidir-llada8b")
+model, report = load_diffusion_prm(
+    checkpoint=path,
+    model_path="GSAI-ML/LLaDA-8B-Base",
+    local_files_only=False,
+)
+```
+
+**Role:** cross-backbone diagnostic; it is not the Dream headline scorer.
+
+Paper: [https://arxiv.org/abs/2609.35472](https://arxiv.org/abs/2609.35472).

adapter.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6d8ee819ff0e474930300bf11d37db9b76c01254183e4cb4c06b28fb8d579a7
+size 51671716

config.json

@@ -0,0 +1,45 @@
+{
+  "base_model": "GSAI-ML/LLaDA-8B-Base",
+  "adapter_type": "LoRA plus reward head",
+  "attention": "bidirectional",
+  "pool_strategy": "mean",
+  "mask_aware_pooling": true,
+  "step_embedding": true,
+  "lora": {
+    "r": 16,
+    "alpha": 32,
+    "dropout": 0.05,
+    "target_modules": [
+      "q_proj",
+      "v_proj"
+    ]
+  },
+  "release_status": "public release",
+  "checkpoint_identity_status": "exact cross-backbone diagnostic checkpoint",
+  "paper_role": "cross-backbone diagnostic; it is not the Dream headline scorer",
+  "adapter_bytes": 51671716,
+  "parameter_prefixes_kept": [
+    "lora_A",
+    "lora_B",
+    "reward_head",
+    "step_proj",
+    "step_embed"
+  ],
+  "causal": false,
+  "no_step_embed": false,
+  "no_mask_aware": false,
+  "lora_r": 16,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
+  "step_embed_dim": 256,
+  "reward_hidden": 1024,
+  "seed": 42,
+  "max_steps": null,
+  "tags": [
+    "dllm-prm-gap",
+    "discrete-diffusion",
+    "gsm8k",
+    "process-reward-model",
+    "llada-8b"
+  ]
+}