不微调模型，用通用+专业双引擎提升医疗问答准确率
TAGS: A Test-Time Generalist-Specialist Framework with Retrieval-Augmented Reasoning and Verification
- 测试时融合通用模型与领域专家模型，协同推理
- 在9个医疗基准上使GPT-4o准确率提升13.8%
- 适合无标注数据场景下快速部署高可靠医疗AI
近期基于思维链的提示方法显著提升了大语言模型在零样本医疗推理中的表现。然而，提示方法往往浅层且不稳定，而微调过的医学大模型在分布偏移下泛化能力差，难以适应未见临床场景。为此，我们提出TAGS，一种无需模型微调或参数更新的测试时框架，通过通用模型与领域专用专家模型的互补视角实现协同推理。为支持该过程，引入两个辅助模块：基于语义和推理路径相似性的分层检索机制，用于获取多尺度示例；以及可靠性评分器，评估推理一致性以指导最终答案聚合。TAGS在九个MedQA基准上表现强劲，使GPT-4o准确率提升13.8%，DeepSeek-R1提升16.8%，并将7B基础模型准确率从14.1%提高至23.9%。结果超越多个微调过的医学大模型，且无需任何参数更新。代码将发布于https://github.com/JianghaoWu/TAGS。
原文摘要 · Abstract (English)
Recent advances such as Chain-of-Thought prompting have significantly improved large language models (LLMs) in zero-shot medical reasoning. However, prompting-based methods often remain shallow and unstable, while fine-tuned medical LLMs suffer from poor generalization under distribution shifts and limited adaptability to unseen clinical scenarios. To address these limitations, we present TAGS, a test-time framework that combines a broadly capable generalist with a domain-specific specialist to offer complementary perspectives without any model fine-tuning or parameter updates. To support this generalist-specialist reasoning process, we introduce two auxiliary modules: a hierarchical retrieval mechanism that provides multi-scale exemplars by selecting examples based on both semantic and rationale-level similarity, and a reliability scorer that evaluates reasoning consistency to guide final answer aggregation. TAGS achieves strong performance across nine MedQA benchmarks, boosting GPT-4o accuracy by 13.8%, DeepSeek-R1 by 16.8%, and improving a vanilla 7B model from 14.1% to 23.9%. These results surpass several fine-tuned medical LLMs, without any parameter updates. The code will be available at https://github.com/JianghaoWu/TAGS.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。