arXiv：cs.AI· Qi Cheng, Shengyu Chen, Wei Cheng, Zhengzhang Chen, Xiaowei Jia, Haoyu Wang, Haifeng Chen·· 6 小时前AI 评分38
WorkflowOps：为多智能体工作流编排学习智能体协作先验
WorkflowOps: Learning Agent Collaboration Priors for Multi-Agent Workflow Orchestration
AI 导读
WorkflowOps 是一个多智能体工作流编排框架，可从历史工作流中学习智能体协作先验，并按需扩展智能体池。它通过转移概率矩阵引导 DAG 构建，借助充分性驱动的智能体创建循环按语义匹配分数补足能力缺口，并用分层语义匹配将 LLM 路由调用减少逾 80%。在代码、数学与问答混合套件上，其端到端通过率优于近期工作流构建基线，结构化可分解任务提升最大。
正文
Abstract:Multi-agent systems are increasingly deployed for complex knowledge work, yet their orchestration layers remain largely memoryless: each new task is decomposed, assigned, and executed from scratch with no benefit from prior successful executions. We present WorkflowOps, a multi-agent workflow orchestration framework that learns agent collaboration priors from historical workflows and expands its agent pool on demand to cover new capability requirements. Our approach introduces three coupled mechanisms. First, a transition probability matrix captures pairwise agent collaboration frequencies from past workflows and applies them as soft guidance during DAG workflow construction through intra-layer ordering optimization, probability-thresholded edge suggestion, and transitive reduction for parallelism maximization. Second, a sufficiency-driven agent creation loop detects capability gaps via semantic matching scores, generates specialized agents through an LLM, and simultaneously injects them into the collaboration matrix, so that newly created agents are immediately usable with predicted collaboration priors. Third, a layered semantic matching strategy uses pre-trained sentence embeddings for fast, deterministic capability matching as a first pass, invoking LLM verification only for low-confidence cases, thereby reducing LLM routing calls by over 80\% compared to pure-LLM approaches. Experiments on mixed code, math, and question-answering suites show that WorkflowOps improves end-to-end pass rates over recent workflow-construction baselines, with the largest gains on structured, decomposable tasks where past agent handoff patterns transfer.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.07860 [cs.AI] |
| (or arXiv:2610.07860v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.07860 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Qi Cheng [view email]
[v1]
Tue, 6 Oct 2026 07:01:38 UTC (77 KB)
来源：arXiv：cs.AI · arxiv.org