提出层级记忆统一理论，解析语言智能体的长上下文处理机制。
Toward a Theory of Hierarchical Memory for Language Agents
- 用三个操作符分解层级记忆：提取、压缩分组、遍历检索
- 发现代表函数自足性影响检索策略选择，形成耦合约束
- 适用于文档、对话、智能体执行等多种场景，覆盖11个现有系统
许多近期的长上下文与智能体系统通过引入层级记忆来应对上下文长度限制：从原始数据中提取原子信息单元，通过分组与压缩构建多层表征，并在令牌预算下遍历该结构以检索内容。尽管实现方式反复出现，但缺乏统一的形式化框架用于比较设计选择。本文提出一个基于三个算子的统一理论：提取（α）将原始数据映射为原子信息单元；粗化（C = (π, ρ)）对单元进行划分并为每组分配代表性单元；遍历（τ）根据查询和预算决定包含哪些单元进入上下文。我们识别出代表函数ρ的自足性谱系，并揭示其如何约束可行的检索策略（即粗化-遍历耦合）。最后，我们在涵盖文档层次、对话记忆和智能体执行轨迹的十一个现有系统上实例化该分解，展示了其普适性。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。