arXiv cs.AI· Yihao Wang, Haoran Xu, Renjie Gu, Yixuan Ye, Xinyi Chen, Xinyu Mu, Yuan Gao, Chunxiao Guo, Peng Wei, Jinjie Gu, Huan Li, Ke Chen, Lidan Shou·· 9 天前AI 评分24
MedMemoryBench：面向个性化医疗智能体的记忆基准测试
MedMemoryBench: Benchmarking Agent Memory in Personalized Healthcare
AI 导读
研究者推出 MedMemoryBench，用于评测个性化医疗智能体在长期临床跟踪中的记忆能力，数据集包含约 2,000 个会话、16,000 轮交互。该基准采用"边构建边评估"的流式评测协议，并系统研究了信息持续涌入导致检索与推理鲁棒性下降的"记忆饱和"现象。测试显示主流架构在复杂医疗推理和噪声鲁棒性上存在严重瓶颈。
正文 · 原文
Authors:Yihao Wang, Haoran Xu, Renjie Gu, Yixuan Ye, Xinyi Chen, Xinyu Mu, Yuan Gao, Chunxiao Guo, Peng Wei, Jinjie Gu, Huan Li, Ke Chen, Lidan Shou
Abstract:The large-scale deployment of personalized healthcare agents demands memory mechanisms that are exceptionally precise, safe, and capable of long-term clinical tracking. However, existing benchmarks primarily focus on daily open-domain conversations, failing to capture the high-stakes complexity of real-world medical applications. Motivated by the stringent production requirements of an industry-leading health management agent serving tens of millions of active users, we introduce MedMemoryBench. We develop a human-agent collaborative pipeline to synthesize highly realistic, long-horizon medical trajectories based on clinically grounded, synthetic patient archetypes. This process yields a massive, expertly validated dataset comprising approximately 2,000 sessions and 16,000 interaction turns. Crucially, MedMemoryBench departs from traditional static evaluations by pioneering an "evaluate-while-constructing" streaming assessment protocol, which precisely mirrors dynamic memory accumulation in production environments. Furthermore, we formalize and systematically investigate the critical phenomenon of memory saturation, where sustained information influx actively degrades retrieval and reasoning robustness. Comprehensive benchmarking reveals severe bottlenecks in mainstream architectures, particularly concerning complex medical reasoning and noise resilience. By exposing these fundamental flaws, MedMemoryBench establishes a vital foundation for developing robust, production-ready medical agents.
| Subjects: | Artificial Intelligence (cs.AI) |
| MSC classes: | 68T07, 68T50 |
| ACM classes: | I.2.7; I.2.1 |
| Cite as: | arXiv:2605.11814 [cs.AI] |
| (or arXiv:2605.11814v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2605.11814 arXiv-issued DOI via DataCite |
Submission history
From: Wang Yihao Cyan [view email]
[v1]
Tue, 12 May 2026 09:06:40 UTC (16,341 KB)
[v2]
Thu, 8 Oct 2026 02:51:38 UTC (16,344 KB)
来源：arXiv cs.AI · arxiv.org