arXiv cs.AI· Jinyang Li, Zichuan Fu, Xian Wu, Wenlin Zhang, Yimin Deng, Kaifeng Guo, Guojing Li, Junjia QI, Xiangyu Zhao·· 13 小时前AI 评分32
ProactMem：面向事件驱动 LLM 智能体的主动记忆机制
Proactive Memory for Event-Driven LLM Agents
AI 导读
研究者提出 Proactive Memory（ProactMem），将历史组织为近期事件、时间轨迹和行为规律三类，轨迹事实对照源事件核验，规律经历史回放评估后再匹配当前条件。团队同时构建纵向基准 ProactiveStream，覆盖三类应用场景，智能体按时间顺序处理孤立事件流。ProactMem 在该基准上取得 52.6% 的 F1，相对五个外部基线中最强者的相对提升为 21.4%，代码已公开。
正文 · 原文
Abstract:Proactive LLM agents anticipate service opportunities from ongoing observations and accumulated history. Existing approaches emphasize recent context or retrieval of related memories, providing limited support for validating recurring behavioral regularities and assessing their applicability to the current event. We propose Proactive Memory (ProactMem), which organizes history into recent events, temporal trajectories, and behavioral regularities. Trajectory facts are checked against source events, and regularities are assessed through historical replay before being matched to current conditions. We also construct ProactiveStream, a longitudinal benchmark covering three application scopes, where agents process isolated event streams chronologically under episode-aware opportunity labels. On ProactiveStream, ProactMem achieves an overall effectiveness (F1) of 52.6%, a 21.4% relative improvement over the strongest of five external baselines. Code is available at: this https URL.
| Comments: | 15 pages |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.36967 [cs.AI] |
| (or arXiv:2610.36967v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36967 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Jinyang Li [view email]
[v1]
Tue, 27 Oct 2026 10:23:50 UTC (499 KB)
来源：arXiv cs.AI · arxiv.org