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
Donghua Zhang · Zizheng Wang · Ze Dong · Yiding Feng · Siyu Qian
中文摘要
面向跨就诊跟踪患者的 LLM 医疗助手需要一种记忆，能在患者临床状态变化时保持其正确。考察五个记忆系统在纵向医疗对话上的错误，发现三类反复出现的原因：检索到的证据给出过时状态、混入他人的健康事实、或缺少答案所依赖的安全事实。MedGraph 是针对这三种情况构建的图记忆。构建阶段从每次会话中抽取带日期的事实，将每条事实归属到其所描述的人，并把用药、诊断和测量记录链接成保留早期状态的链。检索阶段以患者链为单位排序，按问题所问的时间点读取每条链，并补充原始对话文本；患者档案在每个问题中列出所有已记录的安全事实。在 MedMemoryBench 上，MedGraph 在 GPT 与 Qwen 两种骨干下均取得最高总体准确率，分别比五个基线中最强者高 10.5 和 10.4 分，且在 GPT 设置下记忆构建与回答高效。
关键要点
- 01问题：现有记忆系统在纵向医疗对话中易返回过时状态、混入他人事实或遗漏安全事实
- 02方法：构建带日期、按人归属的事实链，检索时按问题时间点读取链并附原始对话
- 03结果：MedMemoryBench 上 GPT 与 Qwen 骨干分别领先最强基线 10.5 和 10.4 分
- 04效率：GPT 设置下记忆构建与回答均高效
解读
尚无解读。
原始英文摘要
arXiv:2611.08433v1 Announce Type: new 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five memory systems on longitudinal medical dialogue and find three recurring causes: the retrieved evidence gives an outdated state, includes another person's health facts, or lacks a safety fact the answer depends on.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.