面向紧急医疗服务问答的专家引导提示与检索增强生成
Expert-Guided Prompting and Retrieval-Augmented Generation for Emergency Medical Service Question Answering
AI总结：
针对紧急医疗服务问答中现有大语言模型未利用领域结构化知识的问题，构建EMSQA数据集，提出Expert-CoT提示策略与ExpertRAG流水线，显著提升模型答题准确率，32B增强模型可通过全部EMS资质模拟考试。
摘要：
大语言模型（LLMs）在医疗问答任务中已展现出应用潜力，但这类模型往往忽略专业人员所依赖的领域专属知识，例如创伤、气道管理等临床学科领域，以及EMT（紧急医疗技术员）、护理人员等对应的资质等级。现有方法通常采用通用提示或检索策略，未利用这类结构化上下文，在高风险场景下的性能受限。针对这一缺口，我们构建了EMSQA数据集，该数据集包含2.43万道选择题，覆盖10个临床学科领域、4个资质等级，同时配套有经整理、与各学科领域对齐的知识库，包含4万份文档、总计200万token。基于EMSQA，我们提出两项方法：一是Expert-CoT提示策略，让思维链（CoT）推理以特定临床学科领域和资质等级为条件；二是ExpertRAG检索增强生成流水线，将回答锚定在与学科对齐的文档及真实患者数据上。在4个大语言模型上开展的实验显示，Expert-CoT相比原生CoT提示准确率最高提升2.05%；此外，Expert-CoT与ExpertRAG结合的方案，相比标准RAG基线准确率最高提升4.59%。值得注意的是，经领域知识增强的32B参数大语言模型通过了全部计算机自适应EMS资质模拟考试。
英文摘要：
Large language models (LLMs) have shown promise in medical question answering, yet they often overlook the domain-specific expertise that professionals depend on, such as the clinical subject areas (e.g., trauma, airway) and the certification level (e.g., EMT, Paramedic). Existing approaches typically apply general-purpose prompting or retrieval strategies without leveraging this structured context, limiting performance in high-stakes settings. We address this gap with EMSQA, an 24.3K-question multiple-choice dataset spanning 10 clinical subject areas and 4 certification levels, accompanied by curated, subject area-aligned knowledge bases (40K documents and 2M tokens). Building on EMSQA, we introduce (i) Expert-CoT, a prompting strategy that conditions chain-of-thought (CoT) reasoning on specific clinical subject area and certification level, and (ii) ExpertRAG, a retrieval-augmented generation pipeline that grounds responses in subject area-aligned documents and real-world patient data. Experiments on 4 LLMs show that Expert-CoT improves up to 2.05% over vanilla CoT prompting. Additionally, combining Expert-CoT with ExpertRAG yields up to a 4.59% accuracy gain over standard RAG baselines. Notably, the 32B expertise-augmented LLMs pass all the computer-adaptive EMS certification simulation exams.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
