--- title: Duan canonical_url: "https://www.modelscope.cn/models/wipenhan/Duan" md_url: "https://www.modelscope.cn/models/wipenhan/Duan.md" repository: wipenhan/Duan chinese_name: Duan last_updated: 2026-09-27 license: cc-by-nc-4.0 pipeline_tag: text-classification tasks: - text-classification base_model: - convaiinnovations/laya-multilingual base_model_relation: finetune parameters: 321.9M tensor_type: - F16 library_name: - safetensors language: - en - zh downloads: 9 stars: 0 tags: - medical - emergency-medicine - triage - decision-support - calibrated-confidence - system-1 --- # Duan > Duan - wipenhan 在 ModelScope 开源的模型。Duan 是一个面向医疗场景的 Decision-Native 模型。它不生成大段文本，而是在单次前向中直接返回结构化决策、各选项的概率分布、置信度。当置信度不足时，Duan 的设计意图是把病例交给人工专家，或交给更具分析能力的 System 2。 wipenhan/Duan 是 ModelScope 魔搭社区上的 321.9M 参数text-classification模型，采用 cc-by-nc-4.0 许可，基于 convaiinnovations/laya-multilingual 构建。 - **Repository**: wipenhan/Duan - **License**: cc-by-nc-4.0 - **Tasks**: text-classification - **Parameters**: 321.9M - **Base model**: convaiinnovations/laya-multilingual - **Tags**: medical, emergency-medicine, triage, decision-support, calibrated-confidence, system-1 - **Downloads**: 9 - **Stars**: 0 - **Last updated**: 2026-09-27 Source: https://www.modelscope.cn/models/wipenhan/Duan --- # Duan **Duan** is a fine-tuned [Laya](https://github.com/convaiinnovations/laya) **System-1 decision engine** for emergency and clinical decision tasks. It answers a question about a patient state in **a single forward pass** — no autoregressive decoding, no chain-of-thought — and returns a typed decision together with a calibrated probability distribution, a confidence estimate, and an abstention signal. ## Architecture | | | |---|---| | Backbone encoder | [`jhu-clsp/mmBERT-base`](https://huggingface.co/jhu-clsp/mmBERT-base) (ModernBERT, 768 hidden, 322 M params) | | Head | 2-layer Transformer encoder + typed decision scorer (`Embedding(3, d)` type bias + `LayerNorm/Linear/GELU/Linear` scorer) | | Sequence budget | `max_len = 1024`, `head_max_len = 256` | | Task format | `[CLS] instructions [SEP] [MASK] opt₀ … [MASK] optₙ [SEP] state [SEP]` | **Three question types**, sharing one scorer — the only structural difference is a 3-row type embedding: | type | `criteria` | `gold` keys | answer | |---|---|---|---| | `choice` | `{label: description}` | same labels | argmax option + distribution | | `score` | `[level 0 desc, level 1 desc, …]` | `"0" … "n-1"` | expected level + distribution | | `noul` | `{"false": …, "true": …}` | `false` / `true` | P(true) | ## Output `Decision · Probability · Confidence · Abstention` — a full distribution over options, an expected level for `score`, `P(true)` for `noul`, plus a calibrated confidence and an action head that can **abstain** rather than guess. ## Training | | | |---|---| | Objective | RLCD — group sampling from perturbed logits → strictly proper scoring reward (log score + spherical, **+ RPS for `score`**) → group-normalised advantage × policy gradient, plus soft-label cross-entropy | | Steps | 12,992 optimizer steps (8 epochs) | | Effective batch | 64 sequences/step | | Data | 103,886 train items + 1,500 held-out for calibration (stratified) | | Hardware | 1 × A800 80 GB, bf16, 5.2 h | | Seeds | 20260925 (train) / 20260922 (calibration split) | ### Training data (8 sources, 3 languages) | Source | Task | Lang | Licence | |---|---|---|---| | NHAMCS 2018–2022 (CDC/NCHS) | ED triage (ESI 1–5) | en | MIT (US public data) | | UCI Diabetes 130-US hospitals | readmission window (score, 3 levels) | en | CC BY 4.0 | | MedQA — US / Mainland / Taiwan | medical MCQ | en / zh-Hans / zh-Hant | see upstream dataset | | CARE-Bench | care escalation (4 levels) | en | **CC BY-NC 4.0** | | DDXPlus (English) | differential-diagnosis re-ranking | en | CC BY 4.0 | | PubMedQA PQA-L | conclusion supported (noul) | en | MIT | ⚠️ No source's raw text is redistributed here — only the resulting weights. Several sources are **not** clinical outcomes: MedQA labels are exam answer keys, CARE-Bench labels are human-annotated consensus grades, PubMedQA labels are literature annotations, and **DDXPlus is fully synthetic**. ## Evaluation (test split, n = 10,538) | Task | n | acc | majority | macro-F1 | ECE | other | |---|---|---|---|---|---|---| | `ddx_diagnosis` | 2000 | 0.9905 | 0.1660 | 0.9900 | 0.0071 | synthetic data | | `ed_triage` (choice, 5) | 2000 | 0.5600 | 0.4975 | 0.3490 | 0.1423 | QWK 0.4290 | | `readmission_window` (score, 3) | 2000 | 0.5810 | 0.5350 | 0.4178 | 0.0391 | QWK 0.2239, MAE 0.4875, ±1 level 93.2 % | | `medqa_mcq` | 4415 | 0.3647 | 0.2258 | 0.3659 | 0.4564 | en 0.2620 / zh-Hans 0.4505 / zh-Hant 0.3220 (acc) | | `care_escalation` | 79 | 0.6456 | 0.3418 | 0.6317 | 0.2951 | small n | | `pubmedqa` (noul) | 44 | 0.7273 | 0.6364 | 0.6966 | 0.2727 | AUROC 0.6708, PR-AUC 0.8135 | **Read these carefully.** Majority-class baselines are given because every task is imbalanced; macro-F1 is the primary metric; cross-task macro-F1 is **not** comparable (different label spaces). `ddx_diagnosis` is high because DDXPlus is synthetic and its option set comes from the generator's own posterior — this is **closed-set re-ranking**, not open-set diagnosis. ## Calibration `rl_agent_config.json` ships a `temperature_by_options` table keyed by `":"` and an additional `temperature_by_qid` table keyed by `"|:"`. The per-qid table exists because the coarse bucket key cannot separate tasks that share a bucket — MedQA (5 options), NHAMCS triage (5 options) and CARE-Bench (4 options) all fall into `choice:3-5`, while their uncalibrated ECE differs by an order of magnitude. To use it, rescale probabilities offline — the model emits `p = softmax(logits / t)`, so `p_new ∝ p^(t_old / t_new)`: ```python def rescale(probs, t_from, t_to): e = t_from / t_to w = {k: max(v, 1e-12) ** e for k, v in probs.items()} s = sum(w.values()) return {k: v / s for k, v in w.items()} ``` Values are clamped to `[0.5, 5]` on load; any out-of-range fitted value is also recorded under `temperature_by_qid_raw`. ## Usage ```python import laya agent = laya.Agent("wipen/Duan") res = agent.predict( state={"age": 67, "chief_complaint": "chest pain", "heart_rate": 118, "spo2": 91, ...}, questions={"acuity": { "type": "choice", "instructions": "What ESI acuity level should this visit be triaged to?", "criteria": {"1": "requires immediate life-saving intervention", ...}, }}, ) print(res["answers"]["acuity"]["choice"], res["answers"]["acuity"]["probability"]) ``` ## Licence Weights released under **CC BY-NC 4.0**, because one training source (CARE-Bench) is licensed CC BY-NC 4.0 and another (DDXPlus) requires attribution. The base model (`convaiinnovations/laya-multilingual`) and the training code are Apache-2.0. **Commercial use is not permitted** under this licence. If you need commercial rights, the CARE-Bench portion must be removed and the model retrained, or separate permission obtained from that dataset's authors. ## Citation ```bibtex @misc{duan2026, title = {Duan: A Healthcare System 1 Decision Model}, author = {Han, Weipeng}, year = {2026}, howpublished = {\url{https://github.com/wipen/Duan}}, note = {Work in progress} } ``` If you use the underlying engine, please also cite Laya: ```bibtex @misc{laya, title = {Laya: fast, non-autoregressive System 1 decision engine}, author = {{ConvAI Innovations}}, howpublished = {\url{https://github.com/convaiinnovations/laya}}, note = {Apache-2.0} } ```