让大模型智能体用状态树管理记忆，更准完成长任务。
Beyond Semantic Organization: Memory as Execution State Management for Long-Horizon Agents
- 用分层状态树记录交互，按执行路径动态组织记忆
- 在MemoryArena上提升成功率7.8~20.4个百分点，节省55.1%令牌
- 适合需要长期规划与错误隔离的复杂智能体系统
基于大模型的智能体越来越多地处理具有依赖性决策的长周期任务，每一步操作都会改变未来约束，中间错误可能持续扩散。现有RAG与智能体记忆系统按语义相似性组织历史记录，仅在决策时检索相关内容。我们指出这种设计与执行状态依赖不匹配：它打散决策轨迹，混入有效与错误轨迹，阻碍状态重建与错误隔离。为此提出MAGE（Memory as Agent-Guided Exploration），一种主动的状态管理机制，将交互存储于分层状态树中。智能体从根节点到当前节点的路径获取状态信息，融合子目标摘要、近期记录和先前分支提示。四种协同操作维护树结构：Grow记录新轨迹，Compress汇总已完成子目标，Maintain验证摘要，Revise恢复目标边界并切换至新分支。该设计在控制上下文增长的同时保持状态完整性，并将错误段落与主路径隔离。在MemoryArena上的实验表明，MAGE相较基线平均任务成功率提升7.8~20.4个百分点，令牌消耗减少55.1%。
原文摘要 · Abstract (English)
LLM-based agents increasingly tackle long-horizon tasks with interdependent decisions, where each action reshapes future constraints and intermediate errors can cascade. Existing RAG and agent memory systems organize histories by semantic similarity, retrieving content-relevant entries at decision time. We argue that this design mismatches execution-state dependencies: it fragments decision trajectories and mixes valid and erroneous traces, hindering coherent state reconstruction and error isolation. We propose MAGE (Memory as Agent-Guided Exploration), an active execution-state manager that stores interactions in a hierarchical state tree. The agent derives its state from the active root-to-current path, combining subgoal summaries, recent traces, and hints from prior branches. Four coupled operations maintain the tree: Grow records new traces, Compress summarizes completed subgoals, Maintain validates summaries, and Revise restores a target boundary and resumes on a new branch. This design bounds context growth while preserving state integrity and isolating flawed segments from the active path. Experiments on MemoryArena show that MAGE improves the average task success rate by 7.8--20.4 pp over baselines, while reducing token consumption by 55.1%.
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。