De-anonymize model card for public release

README.md

@@ -1,44 +1,28 @@
 ---
-license: other
-tags:
-  - sft
-  - expert-qa
-  - full-ft
-  - medical
-  - exp0
-  - tmp
+base_model: LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct
 language:
-  - ko
-  - en
+- ko
+- en
+license: cc-by-nc-4.0
+tags:
+- speech-to-sft
+- emnlp-2026-industry
 ---
 
-# tmp_expertQA_exaone3.5-7.8b_exp0_medical
-
-QA Generator pipeline ablation 학습 모델 — **연구용 임시 스냅샷** (production 용도 아님).
-
-- **Base model**: `LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct`
-- **Method**: Full FT
-- **Condition**: `exp0` (Phase 0 OFF / Phase 2 OFF)
-- **Domain**: medical
-- **Run ID**: `exaone3.5-7.8b_exp0_medical`
-- **Uploaded**: 2026-05-06T06:09:56Z
-
-## Reproduction
-
-```python
-from huggingface_hub import snapshot_download
-local = snapshot_download(repo_id="Flitto/tmp_expertQA_exaone3.5-7.8b_exp0_medical", token="$HF_TOKEN")
-```
+# expertQA_exaone3.5-7.8b_exp0_medical_fullft
 
-LoRA 어댑터인 경우 PEFT로 로드:
+SFT checkpoint from the EMNLP 2026 Industry Track submission *A Factorial Ablation of a Speech-to-SFT Pipeline: Differential Effects on Data Quality and Downstream Transfer*.
 
-```python
-from peft import PeftModel
-from transformers import AutoModelForCausalLM
-base = AutoModelForCausalLM.from_pretrained("LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct")
-model = PeftModel.from_pretrained(base, local)
-```
+| Field | Value |
+|---|---|
+| Pipeline condition | **Exp 0** (baseline (no refinement)) |
+| Domain | medical |
+| Seed | n/a (single seed 42) |
+| Base model | [LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct](https://huggingface.co/LGAI-EXAONE/EXAONE-3.5-7.8B-Instruct) |
+| Training | Full FT (ZeRO-3) |
+| Upstream STT | In-house STT (paper main pipeline) |
+| License | CC BY-NC 4.0 (research and non-commercial use only) |
 
-## License
+**Intended use**: research and non-commercial use only, matching the consent scope of the source audio.
 
-Inherits the base model's license. EXAONE/Llama/Gemma는 각 제공사 라이선스를 따른다.
+**Companion repository** (code, configs, prompts, sample QA): <https://github.com/flitto/speech-to-sft-ablation-paper>