MedLoCoMo: A Long-Context Multi-Session Medical Dialogue Benchmark for Large Language Models

AI 中文解读

新基准MedLoCoMo让AI面临“长期病历”大考，它专门测试大语言模型能否在患者多年多次的就诊对话中理清头绪、准确推理，结果发现，比起只看一次就诊记录，AI要联系前后多次就诊信息时表现明显更差——哪怕模型拥有超长上下文或借助外部记忆检索也难搞定。通俗来说，这就像让AI扮演一位老医生，不仅要看懂病人最近一次问诊时说的“头疼”，还得联想到三年前的“车祸后遗症”和去年的“复查记录”，甚至要能判断哪些问题是根本回答不了的。研究人员从真实医疗数据里整理出100位患者的完整就诊时间线，每条对话平均超过74万个词元，远超现有测试集的难度。这项工作的实际影响在于，它暴露出当前AI在医疗场景下的短板——片段式记忆容易，长期连贯推理难。未来如果能攻克这一关，医疗AI就能更靠谱地辅助医生做全病程管理，比如帮你调阅几年间的检查结果、提醒药物冲突，甚至主动告诉你“这个症状需要立即复诊”，让个性化医疗真正从概念走向现实。

arXiv:2607.22566v1 Announce Type: new Abstract: MedLoCoMo is a Medical Long-Context Memory benchmark for patient-specific clinical reasoning over multi-admission medical dialogue. Existing medical QA benchmarks largely test short context knowledge or single document grounding, leaving open whether LLMs can use, connect, and abstain over longitudinal patient histories. We build MedLoCoMo from deidentified MIMIC-IV and MIMIC-IV-Note records by constructing admission-level clinical packets, synthesizing grounded doctor-patient conversations, and generating evidence linked QA items over single-admission, cross-admission, and adversarial unanswerable settings. The benchmark contains 100 patient timelines averaging 1,669.8 turns, 29.7 sessions, and 74,512.2 tokens per conversation. Across the evaluated baselines, cross-admission reasoning is consistently harder than localized evidence use, even when models have long context windows or use external memory or retrieval methods. The code and MedLoCoMo benchmark release is available at https://github.com/leozzy13/MedLoCoMo for use and reproducibility.