Pay for the Fault, Not the Flow: Label-Free In-Flow Multi-Agent Workflow Optimization
作者: Xuehang Guo, Haoyu Wang, Shengyu Chen, Zach Chen, Wei Cheng, Qingyun Wang, Haifeng Chen
分类: cs.AI, cs.CL
发布日期: 2026-10-01
🔗 代码/项目: PROJECT_PAGE
💡 一句话要点
提出InFlowOp以优化多智能体工作流,解决任务分解与评估问题
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 多智能体系统 工作流优化 无标签学习 任务分解 智能体选择 成本评估 自动化调度
📋 核心要点
- 现有多智能体工作流构建方法在任务分解和智能体选择上存在不确定性,导致优化和评估过程复杂且成本高。
- 本文提出InFlowOp,通过无标签的成本评估机制优化任务分解和智能体分配,提升工作流的执行效率和准确性。
- 实验表明,InFlowOp在多种领域中相较于单智能体基线提升了最高11.97%,并在流动优化中实现了9.64%的提升。
📝 摘要(中文)
大型语言模型(LLMs)在构建多智能体工作流时面临诸多挑战,包括任务如何细分、选择哪个智能体执行子任务以及何时创建新专家等。现有方法在评估工作流成功与否时,通常需要参考答案或训练评估者,且修复过程成本高昂。为此,本文提出InFlowOp,通过一种无标签的成本评估机制,在任务执行前后优化工作流。InFlowOp在执行过程中能够以最低成本修正故障,并通过Braid基准测试验证其在多智能体协调方面的有效性。实验结果显示,InFlowOp在多个领域中相较于单智能体基线提升了最高11.97%。
🔬 方法详解
问题定义:本文旨在解决多智能体工作流构建中的任务分解、智能体选择及评估问题。现有方法通常依赖于参考答案或训练评估者,导致优化过程复杂且成本高昂。
核心思路:InFlowOp通过一种无标签的成本评估机制,动态调整任务分解的粒度和智能体的分配,确保在执行过程中能够以最低成本修正故障。
技术框架:InFlowOp的整体架构包括任务分解模块、智能体选择模块和故障修正模块。在任务执行前,系统根据成本评估确定任务的细分和智能体的分配;在执行过程中,系统实时监控并修正故障。
关键创新:InFlowOp的主要创新在于其无标签成本评估机制,能够在任务执行前后动态优化工作流,与传统方法相比,避免了固定模板的限制。
关键设计:在设计中,InFlowOp使用了一种基于智能体能力与任务需求之间的成本权衡的评估函数,确保每个决策的成本最小化,同时考虑智能体的运行效率。
🖼️ 关键图片
📊 实验亮点
在实验中,InFlowOp在多种任务场景下表现优异,相较于单智能体基线,性能提升最高达11.97%。在流动优化过程中,InFlowOp实现了9.64%的提升,展示了其在多智能体协调能力上的优势。
🎯 应用场景
InFlowOp的研究成果在多个领域具有广泛的应用潜力,包括自动化任务调度、智能客服系统和复杂项目管理等。通过优化多智能体工作流,该方法能够显著提高任务执行的效率和准确性,降低资源消耗,具有重要的实际价值和未来影响。
📄 摘要(原文)
Large language models (LLMs) increasingly construct multi-agent workflows that decompose a complex task and assign specialist agents from a pool. However, building such a workflow well remains challenging: how finely to divide the task, which agent to trust with each subtask, and when to create a new specialist are all critical decisions a workflow constructor needs to settle up front. Thus, whether each subtask succeeds remains unknown until the workflow runs. Yet, improving a workflow is costly. Locating a fault usually requires a reference answer, a graded outcome, or a trained assessor, and the fix is applied to the whole workflow through re-execution, re-search, or retraining. We propose InFlowOp, which prices every decision in one label-free cost that weighs how well an agent's competence meets what a subtask demands against how much that agent takes to run. Before execution, InFlowOp bidirectionally determines the granularity of task decomposition and agent assignment following from the cost rather than from a fixed template. During execution, InFlowOp corrects a fault with the cheapest move via the same cost that serves the workflow both as it is built and as it runs. Facing the workflow-level evaluation challenge, we introduce Braid, a benchmark whose tasks require multi-agent coordination beyond single-agent capability. Across various domains and backbones, InFlowOp outperforms single agent baselines by up to $+11.97\%$, achieving $+9.64\%$ with in-flow optimization. Our project page: https://xhguo7.github.io/InFlowOp/.