解决多智能体协作中记忆混乱与经验无法复用的问题。
StackPlanner: A Centralized Hierarchical Multi-Agent System with Task-Experience Memory Management
- 分层设计，高阶协调与执行分离，显式管理任务级记忆。
- 通过强化学习复用协作经验，显著提升长程任务可靠性。
- 适合复杂知识型任务的多智能体系统研究者使用。
基于大语言模型的多智能体系统，尤其是集中式架构，在处理复杂、知识密集型任务方面展现出巨大潜力。然而，中心智能体常因缺乏记忆管理，导致上下文膨胀、错误累积和跨任务泛化能力差。为解决任务级记忆效率低及协作经验无法复用的问题，本文提出 StackPlanner，一种具有显式记忆控制的分层多智能体框架。该框架通过将高层协调与子任务执行解耦，并结合结构化经验记忆与强化学习，实现对可复用协作经验的检索与利用。在多个深度搜索与智能体系统基准测试中，实验结果表明该方法能有效支持可靠的长程多智能体协作。
原文摘要 · Abstract (English)
Multi-agent systems based on large language models, particularly centralized architectures, have recently shown strong potential for complex and knowledge-intensive tasks. However, central agents often suffer from unstable long-horizon collaboration due to the lack of memory management, leading to context bloat, error accumulation, and poor cross-task generalization. To address both task-level memory inefficiency and the inability to reuse coordination experience, we propose StackPlanner, a hierarchical multi-agent framework with explicit memory control. StackPlanner addresses these challenges by decoupling high-level coordination from subtask execution with active task-level memory control, and by learning to retrieve and exploit reusable coordination experience via structured experience memory and reinforcement learning. Experiments on multiple deep-search and agent system benchmarks demonstrate the effectiveness of our approach in enabling reliable long-horizon multi-agent collaboration.
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。