arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.26252cs.AIcs.DB
收藏

智能体记忆是数据库吗?重新思考长期AI智能体记忆的数据基础

Is Agent Memory a Database? Rethinking Data Foundations for Long-Term AI Agent Memory

Concordia University(康科迪亚大学)

Abdelghny Orogat, Essam Mansour

AI总结:

本文提出将长期AI智能体记忆视为一种新的数据管理工作负载,通过形式化治理演化记忆(GEM)框架,用四个状态级操作替代记录级操作,并论证记录级系统无法满足其正确性条件,最后通过原型MemState验证可行性并指出未来研究方向。

摘要:

长期运行的AI智能体需要持久记忆。记忆支持跨会话的学习,减少重复的上下文注入,并能够审计过去的决策。当前的智能体记忆系统和数据库范式将记忆视为存储。它们将正确性定位在记录、嵌入或边上。每个只提供了长期记忆所需的部分能力。结果导致四种反复出现的故障模式:无节制的增长、缺乏语义修订、容量驱动的遗忘以及只读检索。在我们的愿景中,长期智能体记忆是一种新的数据管理工作负载。其正确性是状态轨迹的属性,而非单个记录的属性。我们将其形式化为治理演化记忆(GEM)。GEM用四个状态级操作替代记录级数据库操作:摄取、修订、遗忘和检索。六个正确性条件控制状态如何演化。三个结构性观察表明,无论存储模型如何,没有记录级系统能够满足这些条件。我们在一个属性图后端上实现了该抽象的原型MemState。MemState验证了可行性并揭示了与原生引擎之间的差距。我们概述了三个研究方向,将记忆中心的数据管理定义为一个工作负载。

英文摘要:

Long-running AI agents need persistent memory. Memory supports learning across sessions, reduces repeated context injection, and enables auditing of past decisions. Current agent memory systems and database paradigms treat memory as storage. They localize correctness at records, embeddings, or edges. Each supplies only some of the capabilities that long-term memory requires. The result is four recurring failure modes: unregulated growth, missing semantic revision, capacity-driven forgetting, and read-only retrieval. In our vision, long-term agent memory is a new data-management workload. Its correctness is a property of the state trajectory, not of individual records. We formalize this as Governed Evolving Memory (GEM). GEM replaces record-level database operations with four state-level operators: ingestion, revision, forgetting, and retrieval. Six correctness conditions govern how the state evolves. Three structural observations establish that no record-level system can satisfy these conditions, regardless of the storage model. We realize the abstraction in MemState, a prototype on a property-graph backend. MemState validates feasibility and exposes the gap to a native engine. We outline three research directions that define memory-centric data management as a workload.

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。

↑