梳理大模型智能体记忆的六大操作与四大研究方向。
Rethinking Memory in LLM based Agents: Representations, Operations, and Emerging Topics
- 区分参数化记忆与上下文记忆，定义六类核心操作。
- 提出四类关键研究主题，涵盖长时、长上下文等挑战。
- 开源数据集与工具，助力记忆机制研究与开发。
记忆是大语言模型（LLM）智能体的核心，但现有综述多关注应用层面（如个性化对话），忽视了记忆动态背后的原子级操作。本文将记忆分为参数化（隐含于模型权重）和上下文型（显式外部数据，结构化/非结构化）两类，并定义六种核心操作：整合、更新、索引、遗忘、检索与压缩。通过映射这些维度，揭示四大研究热点：长时记忆、长上下文处理、参数化修改与多源记忆。该分类体系为记忆相关研究、基准测试与工具提供了结构化视角，明确了大模型智能体中记忆功能的交互机制，推动未来技术发展。相关数据集、论文与工具已公开于 https://github.com/Elvin-Yiming-Du/Survey_Memory_in_AI。
原文摘要 · Abstract (English)
Memory is fundamental to large language model (LLM)-based agents, but existing surveys emphasize application-level use (e.g., personalized dialogue), while overlooking the atomic operations governing memory dynamics. This work categorizes memory into parametric (implicit in model weights) and contextual (explicit external data, structured/unstructured) forms, and defines six core operations: Consolidation, Updating, Indexing, Forgetting, Retrieval, and Condensation. Mapping these dimensions reveals four key research topics: long-term, long-context, parametric modification, and multi-source memory. The taxonomy provides a structured view of memory-related research, benchmarks, and tools, clarifying functional interactions in LLM-based agents and guiding future advancements. The datasets, papers, and tools are publicly available at https://github.com/Elvin-Yiming-Du/Survey_Memory_in_AI.
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。