When Agents Coordinate: Measuring Coordination in Multi-Agent AI Coding· 多智能体 AI 编程时的协作测量
We study how teams of AI coding agents coordinate while solving programming tasks. Current evaluations usually report whether the agents complete the task and how much the run costs, leaving the coordination inside the team largely unmeasured. We introduce an instrument to measure this coordination. Each run is represented as a temporal network in which agents and files are nodes, and messages, file writes, and file reads are timestamped directed edges with an associated cost. We apply this instrument to 1902 runs, each evaluated with a fixed test suite, across configurations that vary the team size, the team structure, and the file policy. The resulting networks show how coordination changes as teams grow and as the work changes. Direct messaging initially increases close to quadratically
研究多智能体 AI 编程协作模式及测量方法。
- 核心方法
- 将每次运行表示为时间网络，其中智能体和文件作为节点，消息传递、文件写入和读取作为带时间戳的有向边，并关联成本。
- 适合谁读
- 研究者、工程师
- 要解决的问题
- 现有评估方法未充分测量多智能体 AI 编程时内部团队协作情况。
- 关键实验
- 应用于 1902 次运行，每次运行使用固定的测试套件，团队规模、结构和文件策略不同，展示了团队协作随团队规模和任务变化的情况。
- 主要贡献
- 提出了一种新的测量多智能体 AI 编程时团队协作的工具。
- 意义与局限
- 意义在于深入理解多智能体协作机制，影响未来多智能体系统的优化设计。局限在于研究主要基于特定编程任务，适用性需进一步验证。