arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.24486cs.AIcs.CL
收藏

AgentFugue:通过集体推理实现长时域任务的智能体扩展

AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning

GSAI, Renmin University of China(GSAI,中国人民大学);Beijing Academy of Artificial Intelligence(北京人工智能研究院)

Yuyang Hu, Hongjin Qian, Shuting Wang, Jiongnan Liu, Tong Zhao, Xiaoxi Li, Zheng Liu, Zhicheng Dou

AI总结:

提出AgentFugue框架,通过共享推理中心实现多个对等智能体并行探索和选择性信息共享,无需显式角色分工或工作流编排,从而提升长时域任务性能。

摘要:

近期长时域智能体任务的进展主要通过更强模型、更好工具和更有效脚手架来扩展单个智能体。相比之下,对于扩展(scaling out)的理解要少得多:多个对等智能体,都针对同一任务,能否在不依赖显式角色分工或工作流编排的情况下成为额外能力来源?我们研究这个问题并提出AgentFugue,一个围绕共享推理中心构建的集体推理框架。当对等智能体并行探索同一任务时,中心记录每个智能体已建立、尝试或排除的简明笔记,并使每个智能体能够以对其当前搜索有用的形式选择性访问其他智能体的发现。这种设计将原本孤立的轨迹转变为可重用中间推理的互联生态,无需集中规划。我们将中心实例化为一个即插即用的通信层,使用监督微调和端到端强化学习进行训练。在我们研究的具有挑战性的长时域设置中,AgentFugue优于强基线。我们的结果表明,集体推理可以将对等智能体系统的扩展转变为能力增益的独特来源,而不仅仅是消耗更多计算的方式。

英文摘要:

Recent progress on long-horizon agentic tasks has been driven largely by scaling up individual agents through stronger models, better tools, and more effective scaffolding. In contrast, much less is understood about scaling out: whether multiple peer agents, all targeting the same task, can become an additional source of capability without relying on explicit role specialization or workflow orchestration. We study this question and propose AgentFugue, a collective reasoning framework built around a shared reasoning hub. As peer agents explore the same task in parallel, the hub records concise notes on what each agent has established, attempted, or ruled out, and enables each agent to selectively access what other agents have discovered in a form useful for its current search. This design turns otherwise isolated trajectories into a connected ecology of reusable intermediate reasoning without requiring centralized planning. We instantiate the hub as a plug-in communication layer, trained with supervised fine-tuning and end-to-end reinforcement learning. Across the challenging long-horizon settings we study, AgentFugue improves over strong baselines. Our results suggest that collective reasoning can turn scaling out peer agent systems into a distinct source of capability gains, rather than merely a way of spending more compute.

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。

↑