arXiv cs.CL· Donghua Zhang, Zizheng Wang, Ze Dong, Yiding Feng, Siyu Qian·· 11 小时前AI 评分34
MedGraph：用图记忆维护长期医疗对话中的患者状态
MedGraph: Maintaining Patient State with Graph Memory for Long-Term Medical Dialogue
AI 导读
MedGraph 是面向长期医疗对话的图记忆系统，通过提取带日期的患者事实、按人归属并链接成用药/诊断/测量链，在检索时按问题所问时间点读取链条并补充原始对话文本。在 MedMemoryBench 上，MedGraph 分别以 GPT 和 Qwen 为底座取得最高总体准确率，较五个基线中最强者高出 10.5 和 10.4 分，且在 GPT 设置下记忆构建与回答效率较高。
正文 · 原文
Abstract:LLM-based medical assistants that follow a patient across visits need a memory that keeps the patient's clinical state correct as it changes. We examine the errors of five memory systems on longitudinal medical dialogue and find three recurring causes: the retrieved evidence gives an outdated state, includes another person's health facts, or lacks a safety fact the answer depends on. We present MedGraph, a graph memory built for these three cases. Construction extracts dated facts from each session, attributes each fact to the person it describes, and links medication, diagnosis, and measurement records into chains that keep earlier states. Retrieval ranks the patient's chains as units, reads each chain at the time the question asks about, and adds the original dialogue text; a patient profile lists every recorded safety fact with every question. On MedMemoryBench, MedGraph has the highest overall accuracy with both a GPT and a Qwen backbone, 10.5 and 10.4 points above the strongest of five baselines, with efficient memory construction and answering in the GPT setting.
| Comments: | 14 pages, 5 figures, 8 tables |
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.08433 [cs.CL] |
| (or arXiv:2611.08433v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08433 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Donghua Zhang [view email]
[v1]
Thu, 5 Nov 2026 18:27:20 UTC (422 KB)
来源：arXiv cs.CL · arxiv.org