ReLMem：学习循环记忆用于纵向电子健康记录建模

中文摘要 AI 辅助

纵向电子健康记录（EHR）建模需要将新的就诊记录与不断扩展的患者历史信息进行整合。然而，临床信息的持续累积使得大型语言模型（LLMs）在处理和保留完整患者历史时面临不断增加的计算和内存成本。一种实用的替代方案是逐次就诊的循环压缩，即将每次新就诊信息整合到一个紧凑且持续更新的患者记忆中。然而，在固定的内存预算下，连续更新必须整合新信息，同时不能逐步丢失后续任务所需的关键历史证据。为了应对这一挑战，我们引入了循环纵向记忆（ReLMem），这是一个学习维护固定容量患者记忆的框架，用于在冻结的LLM下进行高效的下游预测。ReLMem为该LLM配备了轻量级压缩适配器，以从其先前状态和每次新就诊中循环更新记忆，而无需重新读取早期记录。具体而言，我们开发了一种多粒度优化策略，以在循环更新过程中保留与任务相关的信息，并支持从最终记忆进行下游预测。中间监督在相同查询下对齐压缩记忆和完整历史的注意力输出，而预测监督则基于最终记忆，以最小化与真实答案的交叉熵。在基于EHR的药物预测中，ReLMem在将平均保留历史存储减少97.1%的同时，接近了完整历史基线的F1分数。在相同内存预算下，它相较于最强的压缩记忆基线，宏F1和微F1分别提高了4.66和4.75个百分点。这些结果凸显了学习循环患者记忆对于高效纵向EHR建模的价值。

英文摘要

Longitudinal electronic health record (EHR) modeling requires integrating new visits with an expanding patient history. Yet the continual accumulation of clinical information imposes increasing computational and memory costs on large language models (LLMs) when they process and retain complete patient histories. A practical alternative is visit-wise recurrent compression, which incorporates each incoming visit into a compact, continually updated patient memory. However, under a fixed memory budget, successive updates must integrate new information without progressively losing critical historical evidence needed to subsequent tasks. To address this challenge, we introduce Recurrent Longitudinal Memory (ReLMem), a framework that learns to maintain fixed-capacity patient memory for efficient downstream prediction with a frozen LLM. ReLMem equips this LLM with lightweight compression adapters to recurrently update the memory from its previous state and each incoming visit, without rereading earlier records. Specifically, we develop a multi-granularity optimization strategy to preserve task-relevant information throughout recurrent updates and support downstream prediction from the final memory. The intermediate supervision aligns attention outputs from compressed memory and the full history under identical queries, while prediction supervision minimizes cross-entropy with ground truth answers conditioned on the final memory. On EHR-based medication prediction, ReLMem approaches the F1 scores of full-history baseline while reducing average retained historical storage by 97.1%. Under the same memory budget, it improves macro- and micro-F1 over the strongest compressed-memory baseline by 4.66 and 4.75 percentage points, respectively. These results highlight the value of learning recurrent patient memory for efficient longitudinal EHR modeling.

发表机构

• Zhejiang University（浙江大学）
• Tencent Jarvis Lab（腾讯杰维斯实验室）

机构由 AI 辅助整理，请以论文原文为准。