提出动态检索数量的策略，让医疗问答更准确。
heiDS at ArchEHR-QA 2025: From Fixed-k to Query-dependent-k for Retrieval Augmented Generation
- 根据问题自动调整检索数量，而非固定数量
- 实验表明动态策略提升答案事实性和相关性
- 适合医疗问答、需要精准证据支持的场景
本文介绍了我们团队在 ArchEHR-QA 2025 共享任务中的方法 heiDS。该方法基于检索增强生成（RAG）框架，旨在针对患者特定问题，从电子健康记录（EHRs）中提取临床证据生成答案。我们探索了 RAG 框架中的多个组件，重点关注排序列表截断（RLT）策略与溯源方法。不同于传统的固定 top-k RLT 策略，我们采用查询依赖的 k 值检索策略，包括现有的 surprise 与 autocut 方法，以及本文提出的两种新方法 autocut* 与 elbow。实验结果表明，相较于固定-k 策略，该策略在生成事实准确且相关的答案方面具有明显优势。
原文摘要 · Abstract (English)
This paper presents the approach of our team called heiDS for the ArchEHR-QA 2025 shared task. A pipeline using a retrieval augmented generation (RAG) framework is designed to generate answers that are attributed to clinical evidence from the electronic health records (EHRs) of patients in response to patient-specific questions. We explored various components of a RAG framework, focusing on ranked list truncation (RLT) retrieval strategies and attribution approaches. Instead of using a fixed top-k RLT retrieval strategy, we employ a query-dependent-k retrieval strategy, including the existing surprise and autocut methods and two new methods proposed in this work, autocut* and elbow. The experimental results show the benefits of our strategy in producing factual and relevant answers when compared to a fixed-$k$.
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。