arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.02807cs.AI
收藏

群体研究:编排编码智能体以进行开放式发现

SwarmResearch: Orchestrating Coding Agents for Open-Ended Discovery

University of Illinois Urbana-Champaign(伊利诺伊大学厄巴纳-香槟分校)

Yuvraj Virk, Zack Edds, Chunqiu Steven Xia, Lingming Zhang

AI总结:

研究长期运行的编码智能体存在的问题,假设两个设计选择导致该问题,介绍SwarmResearch,一种编排器-子智能体架构,在开放式优化任务中表现良好,其编排器引导扩展能发现更好方案。

摘要:

诸如自动研究之类的长期运行的编码智能体可以持续为开放式问题发现优化方法。然而,它们倾向于收敛到单一的高级方法,然后进行低级编辑,而错过其他更好的解决问题的方法。我们假设两个 harness 级别的设计选择导致了这种行为:在单个长期运行的智能体中积累上下文,并且只暴露单个程序状态进行编辑。我们引入了 SwarmResearch,这是一种编排器-子智能体架构,其中一个 Shepherd 智能体使用全局上下文来引导一群 Search 智能体,每个智能体在其各自的 git 分支中使用局部上下文进行操作。在开放式优化任务中,SwarmResearch 在 13/15 的任务中发现了比现有最先进的 LLM 引导进化和多智能体技术更好或相当的解决方案,这是由更高层次的探索驱动的。与串行和并行智能体的固定扩展相比,SwarmResearch 的编排器引导扩展通过在不同搜索深度调整并行度来发现性能更好的解决方案。

英文摘要:

Long-running coding agents such as autoresearch can persistently discover optimizations for open-ended problems. However, they tend to converge onto a single high-level approach, then proceed with low-level edits while missing other superior approaches to the problem. We hypothesize two harness-level design choices contribute to this behavior: accumulating context in a single long-running agent and only exposing a single program state to edit. We introduce SwarmResearch, an orchestrator-subagent harness in which a Shepherd Agent uses global context to steer a population of Search Agents, each operating with local context in their respective git branch. On open-ended optimization tasks, SwarmResearch discovers better or comparable solutions to state-of-the-art LLM-guided evolution and multi-agent techniques on 13/15 tasks, driven by higher-level exploration. Compared with fixed scaling of serial and parallel agents, SwarmResearch's orchestrator-guided scaling discovers better-performing solutions by adapting parallelism at different search depths.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑