面向去中心化多智能体系统的以责任为中心协调机制
原标题：A Responsibility-Centered Coordination Mechanism for Decentralized Multi-Agent Systems
AI 摘要
论文提出 RCM（以责任为中心的多智能体协调机制），用承诺（commitment）作为协调基本单元，将任务委派与双方明确责任绑定。通过递归委派，智能体在本地编排下游承诺并保留上游责任，配合责任约束决策与责任导向通信协议 RCP，使全局协调无需中心编排即可涌现。在推理、编码和可执行智能体环境中，RCM 任务性能最高提升 18.7 分、责任指标提升 10.2 分，并改善失败与危害检测。
正文节选
A Responsibility-Centered Coordination Mechanism for Decentralized Multi-Agent Systems Abstract Centralized orchestration provides comprehensive multi-agent coordination but creates bottlenecks in scalability, adaptability, and collective performance. Decentralized approaches alleviate these bottlenecks by distributing coordination across agents, yet often leave responsibility for delegated dependencies and failure recovery unclear. To address this gap, we propose RCM, a responsibility-centered