AI TrackerFrontier desk · AEST
Back to news
normalaitech

MedGraph: Maintaining Patient State with Graph Memory for Long-Term Medical Dialogue

arXiv cs.AI · 6 Nov 2026, 04:27 AEST

MedGraph is a new graph memory system designed to maintain accurate patient states during long-term medical dialogues. It addresses common errors found in existing memory systems by linking relevant medical information and ensuring timely retrieval of patient data.

Key facts

Read aloud

MedGraph is a new graph memory system designed to maintain accurate patient states during long-term medical dialogues. It addresses common errors found in existing memory systems by linking relevant medical information and ensuring timely retrieval of patient data. L L M-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five memory systems on longitudinal medical dialogue and find three recurring causes: the retrieved evidence gives an outdated state, includes another person's health facts, or lacks a safety fact the answer depends on.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. On MedMemoryBench, MedGraph has the highest overall accuracy with both a GPT and a Qwen backbone, ten point five and ten point four points above the strongest of five baselines, with efficient memory construction and answering in the GPT setting.

Sources in this cluster