用双演化记忆提升多智能体规划能力，让AI像人一样记规则、改方案。
EvoMem: Improving Multi-Agent Planning with Dual-Evolving Memory
- 设计双记忆机制：任务规则库固定于查询内，反馈信息随迭代动态更新。
- 在行程、会议和日程规划任务中，相比基线模型准确率提升12%-18%。
- 适合需要长期推理与纠错的复杂规划场景，如智能助手或自动化决策。
规划是人工智能解决复杂问题的核心能力，基于大模型的多智能体框架近期扩展了这一能力。然而，人类式记忆在其中的作用仍待探索。理解智能体如何通过记忆协同，对自然语言规划至关重要——迭代推理、约束追踪与错误修正共同决定成败。受认知心理学工作记忆模型启发，我们提出EvoMem，一种基于双演化记忆机制的多智能体框架。该框架包含三个智能体（约束提取器、验证器、执行者）和两个记忆模块：约束记忆（CMem），在跨查询过程中演进，存储任务特定规则与约束，但在单个查询内保持不变；查询反馈记忆（QMem），在单个查询内随迭代累积反馈，用于方案优化。两个记忆模块在每轮查询结束后重置。在行程规划、会议安排和日历调度任务上的评估显示持续性能提升，验证了记忆机制在增强多智能体规划中的有效性。
原文摘要 · Abstract (English)
Planning has been a cornerstone of artificial intelligence for solving complex problems, and recent progress in LLM-based multi-agent frameworks have begun to extend this capability. However, the role of human-like memory within these frameworks remains largely unexplored. Understanding how agents coordinate through memory is critical for natural language planning, where iterative reasoning, constraint tracking, and error correction drive the success. Inspired by working memory model in cognitive psychology, we present EvoMem, a multi-agent framework built on a dual-evolving memory mechanism. The framework consists of three agents (Constraint Extractor, Verifier, and Actor) and two memory modules: Constraint Memory (CMem), which evolves across queries by storing task-specific rules and constraints while remains fixed within a query, and Query-feedback Memory (QMem), which evolves within a query by accumulating feedback across iterations for solution refinement. Both memory modules are reset at the end of each query session. Evaluations on trip planning, meeting planning, and calendar scheduling show consistent performance improvements, highlighting the effectiveness of EvoMem. This success underscores the importance of memory in enhancing multi-agent planning.
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。