Skip to content
Hot storiesContinuously updated

研究团队推出无编排器多智能体代码协同基准ParallACT

1 reports1 reporting sourcesUpdated 17 hours ago

Start here

AI overview

研究团队推出针对无中心编排器的多智能体代码协作评测基准 ParallACT,专门用于测试多个 AI 智能体在共享代码库中处理互补与冲突任务的表现。测试结果显示,并行多智能体的实际表现远逊于集中式单智能体系统。在冲突任务场景下,Claude Code 搭配 Claude Opus 4.8 的任务成功率从 62% 骤降至 16%。研究进一步表明,仅通过告知同伴存在或开放通信通道,难以有效弥合两者间的性能差距,同时还会产生额外的 token 成本。

AI-generated from reports · Updated 13 hours ago

Report timeline

Follow the reports to see different sides of the story.

Nov 3
  1. arXiv 人工智能
    研究团队推出无编排器多智能体代码协同基准 ParallACT

    研究团队推出针对无中心编排器的多智能体代码协作基准 ParallACT,专门测试多个 AI 智能体在共享代码库中处理互补与冲突任务的表现。测试显示并行多智能体表现远逊于集中式单智能体,在冲突任务中 Claude Code 搭配 Claude Opus 4.8 的成功率从 62% 骤降至 16%;仅告知同伴存在或开放通信通道难以有效弥合性能差距,并伴随额外的 token 成本。

Story heat trend

Current heat 6·Comparable peak 9(Nov 3 18:00)·Comparable change, last 24 hours–

02.557.510Nov3Nov3Nov4Nov4

Trends compare only the same parties observed continuously, so the scope can be narrower than current heat counts. Move the pointer or tap the chart to see hourly heat; use the left and right arrow keys on a keyboard.