AutoSwarm：自动化智能体集群的多智能体框架

AI 摘要

香港大学数据科学实验室（HKUDS）提出多智能体框架 AutoSwarm，旨在解决多智能体协作中职责划分不清与依赖关系导致局部错误扩散的问题。该框架包含协作形成与执行协调两个组件，前者界定责任边界与结果验收条件，后者通过共享状态管理依赖与监督关系。在长程编程、交互式规划和金融分析实验中，AutoSwarm 优于单智能体基线和现有多智能体框架，代码已在 GitHub 开源。

正文节选

AutoSwarm: Automating Agent Swarms Abstract Large language model (LLM) agents increasingly automate complex tasks, with multi-agent systems seeking further performance gains through specialization and collaboration. However, distributing work across agents does not automatically yield complementary contributions, and inter-agent dependencies can turn local errors into task failure. To address these challenges, we propose AutoSwarm, a multi-agent framework for collaborative problem solving. Colla

发布时间：2026-10-27 12:00

抓取时间：2026-10-27 12:27

来源机构：arXiv