arXiv cs.AI· Yiran Zeng, Giacomo Frisoni, Lorenzo Molfetta, Gianluca Moro·· 12 小时前AI 评分31
把 UMLS 注入语言模型做医疗问答，真的做完了吗？
Are We Done with UMLS Injection into Language Models for Medical Question Answering?
AI 导读
一篇综述对将 UMLS 知识图谱注入语言模型用于医疗问答（MQA）的现有工作提出质疑：这类注入通常只带来有限的性能提升。作者逐一拆解所选 UMLS 来源与使用方式，发现此前从未讨论的方法论问题，包括 UMLS 覆盖率低和子图检索流程薄弱，并据此给出可操作建议。
正文
Abstract:Maybe not. The Unified Medical Language System (UMLS) has become an oracle for millions of clinicians and health system developers worldwide. Researchers have increasingly used knowledge graphs derived from UMLS to enhance language models' factual accuracy and reasoning traceability in high-stakes tasks. Medical Question Answering (MQA) is a prime testbed for integration, with benchmarks usually involving professional exercises such as disease diagnosis and drug prescription from patient history. Despite the assumed benefits, empirical evidence shows that UMLS-based graph injection typically yields modest performance gains. This paper reviews existing MQA work, critically analyzing the selected UMLS sources and their usage with language models, and probes each of these choices in isolation. We uncover methodological issues never discussed in prior work, from low UMLS coverage to weak subgraph retrieval pipelines. Based on our findings, we provide actionable recommendations to unlock the untapped potential of UMLS as an external knowledge source.
| Comments: | 20 pages, 10 figures, 13 tables. Code and literature annotations: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33801 [cs.AI] |
| (or arXiv:2610.33801v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33801 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Lorenzo Molfetta [view email]
[v1]
Sat, 24 Oct 2026 17:06:58 UTC (1,014 KB)
来源：arXiv cs.AI · arxiv.org