arXiv cs.CL· Zerui Chen, Qinggang Zhang, Qichuan Liu, Zhishang Xiang, Zhimin Wei, Yunbo Tang, Zhuo Li, Jinsong Su, Zhihong Zhang·· 11 小时前AI 评分31
MAC-Bench：通过互联任务评测多智能体协作
MAC-Bench: Benchmarking Multi-Agent Collaboration via Interconnected Tasks
AI 导读
MAC-Bench 是一个通过互联任务评测多智能体协作的基准，不预设智能体角色或通信协议，而是把协作需求嵌入任务依赖与共享资源约束中。它从选定基准任务中挖掘局部协作模式与全局任务结构，经粗到细组合重组成任务图，并在可执行环境中实例化，支持对部分进展和完整任务解决的自动验证。
正文 · 原文
Abstract:Large language model (LLM)-based multi-agent systems (MAS) tackle challenging tasks by dividing work among agents and coordinating their contributions toward shared objectives. Evaluating these systems therefore requires benchmarks that assess not only task-solving performance but also the ability to handle interdependent work. However, existing benchmarks often emphasize task difficulty without substantial collaboration requirements or evaluate only a limited range of predefined collaboration patterns. In this paper, we introduce MAC-Bench, a comprehensive benchmark for evaluating multi-agent collaboration through interconnected tasks. Rather than imposing predefined agent roles or communication protocols, MAC-Bench embeds collaboration requirements directly into task dependencies and shared-resource constraints. Specifically, we mine reusable collaboration structures from selected benchmark tasks at two levels. Local collaboration patterns capture recurring interactions among subtasks, while global task structures characterize how these interactions contribute to complete workflows. Through coarse-to-fine composition, these structures are recombined into task graphs combining different collaboration requirements. Tasks are instantiated in executable environments with automated verification of partial progress and complete task resolution.
| Subjects: | Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.01728 [cs.MA] |
| (or arXiv:2611.01728v1 [cs.MA] for this version) | |
| https://doi.org/10.48550/arXiv.2611.01728 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zerui Chen [view email]
[v1]
Sun, 1 Nov 2026 09:24:22 UTC (2,920 KB)
来源：arXiv cs.CL · arxiv.org