发表机构
The University of Sheffield（谢菲尔德大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
BusMA提出一种基于总线架构的多智能体通信框架，通过共享信道和四种通信意图增强智能体自主性，在13项任务上优于现有层级和路由方法。
AI 中文摘要
多智能体（MA）系统在解决需要规划、工具使用以及综合多来源证据的复杂任务方面表现出色。现有系统通常采用层级式管理者-工作者（HMW）或基于路由的消息传递（RMP）结构作为其通信协议。然而，这些设计限制了智能体的自主性：工作者智能体无法直接咨询特定的“同行”，且错误路由的消息可能传播错误。受计算机系统中总线架构的启发，我们提出了BusMA，一种通信框架，允许任何智能体通过共享信道（即总线）寻址其他智能体。它由智能体注册、消息路由和共享内存管理组件组成。每个配备工具的工作者智能体拥有自己的本地内存，能够推理、行动（使用工具），并通过发布带有特定意图的共享消息进行通信。我们引入了四种意图：讨论、挑战、指导和请求解释，以支持智能体间的细粒度通信。一个主席智能体监控共享内存以协调交互并促进工作者之间的收敛。为评估BusMA的有效性，我们使用两个前沿大语言模型在涵盖视觉推理、数学推理和知识检索的13项任务上进行了广泛实验，结果表明BusMA始终优于最先进的HMW和RMP方法。
英文摘要
Multi-Agent (MA) systems are effective at solving complex tasks that demand planning, tool use, and the synthesis of evidence from multiple sources. Existing systems typically adopt Hierarchical Manager-Worker (HMW) or Router-based Message Passing (RMP) structures as their communication protocol. However, these designs restrict agent autonomy: Worker agents cannot directly consult specific "peers", and misrouted messages can propagate errors. Inspired by bus architectures in computer systems, we propose BusMA, a communication framework that allows any agent to address other agents through a shared channel, i.e., the Bus. It consists of agent registration, message routing, and shared memory management components. Worker agents, each equipped with tools, have their own local memory and can reason, act (tool usage), and communicate by posting shared messages with specific intents. We introduce four intents: discussion, challenge, guidance, and request for explanation, which support fine-grained communication among agents. A Chair agent monitors the shared memory to coordinate interactions and facilitate convergence among Workers. To evaluate the effectiveness of BusMA, we conduct extensive experiments with two frontier LLMs across 13 tasks spanning visual reasoning, mathematical reasoning, and knowledge retrieval. The results demonstrate that BusMA consistently outperforms state-of-the-art HMW and RMP methods.
CommentsCamera-ready version accepted to AACL 2026. 23 pages
