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 2603.21564cs.IRcs.AIcs.ITcs.SImath.IT
收藏

迈向语言智能体的分层记忆理论

Toward a Theory of Hierarchical Memory for Language Agents

University of Alberta(阿尔伯塔大学);York University(约克大学)

Yashar Talebirad, Ali Parsaee, Csongor Y. Szepesvari, Amirhossein Nadiri, Osmar Zaiane

AI总结:

本文提出分层记忆理论,通过三个操作符统一描述语言智能体的上下文处理机制,分析代表函数的自足谱并展示其在不同系统中的应用。

摘要:

许多近期长上下文和智能体系统通过添加分层记忆来解决上下文长度限制问题:它们从原始数据中提取原子单元,通过分组和压缩构建多级代表,并在令牌预算下遍历此结构以检索内容。尽管有重复的实现,但缺乏比较设计选择的共享形式化方法。我们提出一个统一的理论,以三个操作符来描述:提取(α)将原始数据映射到原子信息单元;粗化(C = (π, ρ))将单元分组并为每个组分配代表;遍历(τ)在查询和预算下选择包含在上下文中的单元。我们确定了代表函数ρ的自足谱,并展示了它如何限制可行的检索策略(粗化-遍历耦合)。最后,我们将在文档层次、对话记忆和智能体执行轨迹等十一个现有系统上实例化分解,展示其通用性。

英文摘要:

Many recent long-context and agentic systems address context-length limitations by adding hierarchical memory: they extract atomic units from raw data, build multi-level representatives by grouping and compression, and traverse this structure to retrieve content under a token budget. Despite recurring implementations, there is no shared formalism for comparing design choices. We propose a unifying theory in terms of three operators. Extraction ($α$) maps raw data to atomic information units; coarsening ($C = (π, ρ)$) partitions units and assigns a representative to each group; and traversal ($τ$) selects which units to include in context given a query and budget. We identify a self-sufficiency spectrum for the representative function $ρ$ and show how it constrains viable retrieval strategies (a coarsening-traversal coupling). Finally, we instantiate the decomposition on eleven existing systems spanning document hierarchies, conversational memory, and agent execution traces, showcasing its generality.

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。

↑