用疾病知识图谱提升医疗大模型推理准确率，关键在匹配而非堆砌。
Domain-Specific Knowledge Graphs in RAG-Enhanced Healthcare LLMs
- 构建三种疾病知识图谱，按问题范围精准匹配检索来源。
- 匹配图谱使小中型模型准确率显著提升，盲目拼接反增干扰。
- 大模型依赖自身知识，小模型更需高质量外部知识支持。
大型语言模型生成流畅但常缺乏可信的领域特定推理能力。本文评估领域知识图谱（KG）对医疗领域检索增强生成（RAG）的改进效果，基于PubMed构建三个图谱：$$\mathbb{G}_1$（T2DM）、$$\mathbb{G}_2$（阿尔茨海默病）、$$\mathbb{G}_3$（AD+T2DM）。设计两类探测任务：探针1关注合并的阿尔茨海默病与糖尿病知识，探针2关注$$\mathbb{G}_1$与$$\mathbb{G}_2$的交集。测试七种指令微调的LLM，在六种检索源（无RAG、$$\mathbb{G}_1$、$$\mathbb{G}_2$、$$\mathbb{G}_1$+$$\mathbb{G}_2$、$$\mathbb{G}_3$、$$\mathbb{G}_1$+$$\mathbb{G}_2$+$$\mathbb{G}_3$）及三种解码温度下进行评估。结果表明，探测任务与知识图谱范围的匹配度是决定性因素：精确匹配的检索（尤其是$$\mathbb{G}_2$）带来最一致的性能提升；而随意拼接图谱常引入干扰项，降低准确率。大模型在探针1上常表现优于或持平于无RAG基线，显示其强大的参数先验；小中型模型则更受益于范围明确的检索。温度影响较小，高值通常无效。结论为：应优先采用精准匹配的知识图谱，而非广度整合；并提出图谱选择、模型规模与检索/重排序的实用指南。代码与数据已公开。
原文摘要 · Abstract (English)
Large Language Models (LLMs) generate fluent answers but can struggle with trustworthy, domain-specific reasoning. We evaluate whether domain knowledge graphs (KGs) improve Retrieval-Augmented Generation (RAG) for healthcare by constructing three PubMed-derived graphs: $\mathbb{G}_1$ (T2DM), $\mathbb{G}_2$ (Alzheimer's disease), and $\mathbb{G}_3$ (AD+T2DM). We design two probes: Probe 1 targets merged AD T2DM knowledge, while Probe 2 targets the intersection of $\mathbb{G}_1$ and $\mathbb{G}_2$. Seven instruction-tuned LLMs are tested across retrieval sources {No-RAG, $\mathbb{G}_1$, $\mathbb{G}_2$, $\mathbb{G}_1$ + $\mathbb{G}_2$, $\mathbb{G}_3$, $\mathbb{G}_1$+$\mathbb{G}_2$ + $\mathbb{G}_3$} and three decoding temperatures. Results show that scope alignment between probe and KG is decisive: precise, scope-matched retrieval (notably $\mathbb{G}_2$) yields the most consistent gains, whereas indiscriminate graph unions often introduce distractors that reduce accuracy. Larger models frequently match or exceed KG-RAG with a No-RAG baseline on Probe 1, indicating strong parametric priors, whereas smaller/mid-sized models benefit more from well-scoped retrieval. Temperature plays a secondary role; higher values rarely help. We conclude that precision-first, scope-matched KG-RAG is preferable to breadth-first unions, and we outline practical guidelines for graph selection, model sizing, and retrieval/reranking. Code and Data available here - https://github.com/sydneyanuyah/RAGComparison
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。