arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2511.13593cs.CL
收藏

O-Mem:面向个性化、长时域、自演化代理的全方位记忆系统

O-Mem: Omni Memory System for Personalized, Long Horizon, Self-Evolving Agents

Piaohong Wang, Motong Tian, Jiaxian Li, Yuan Liang, Yuqing Wang, Qianben Chen, Tiannan Wang, Zhicong Lu, Jiawei Ma, Yuchen Eleanor Jiang, Wangchunshu Zhou

AI总结:

O-Mem通过主动用户画像动态提取和更新用户特征,提升个性化代理在复杂环境中的长期交互能力,实现更连贯的响应并优于现有最先进方法。

摘要:

近期基于大语言模型(LLM)的代理在生成类人响应方面展现了显著潜力;然而,它们在复杂环境中维持长期交互仍面临挑战,主要由于上下文一致性及动态个性化方面的限制。现有记忆系统通常依赖语义分组后再检索,这可能导致忽略语义无关但关键的用户信息并引入检索噪声。在本报告中,我们提出了O-Mem的初始设计,这是一种基于主动用户画像的新记忆框架,能够动态提取和更新用户特征及事件记录,基于其主动与代理交互的行为。O-Mem支持对人设属性和主题相关上下文的分层检索,从而实现更适应性和连贯性的个性化响应。O-Mem在公开的LoCoMo基准上达到51.67%,比LangMem(先前的最先进水平)提高了近3%,并在PERSONAMEM上达到62.99%,比A-Mem(先前的最先进水平)提高了3.5%。此外,O-Mem在提升token和交互响应时间效率方面也优于先前的记忆框架。我们的工作为未来开发高效且类人化的个性化AI助手开辟了有前景的方向。

英文摘要:

Recent advancements in LLM-powered agents have demonstrated significant potential in generating human-like responses; however, they continue to face challenges in maintaining long-term interactions within complex environments, primarily due to limitations in contextual consistency and dynamic personalization. Existing memory systems often depend on semantic grouping prior to retrieval, which can overlook semantically irrelevant yet critical user information and introduce retrieval noise. In this report, we propose the initial design of O-Mem, a novel memory framework based on active user profiling that dynamically extracts and updates user characteristics and event records from their proactive interactions with agents. O-Mem supports hierarchical retrieval of persona attributes and topic-related context, enabling more adaptive and coherent personalized responses. O-Mem achieves 51.67% on the public LoCoMo benchmark, a nearly 3% improvement upon LangMem,the previous state-of-the-art, and it achieves 62.99% on PERSONAMEM, a 3.5% improvement upon A-Mem,the previous state-of-the-art. O-Mem also boosts token and interaction response time efficiency compared to previous memory frameworks. Our work opens up promising directions for developing efficient and human-like personalized AI assistants in the future.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑