arXiv cs.CL· Zichuan Fu, Junjia Qi, Xian Wu, Wenlin Zhang, Yimin Deng, Kaifeng Guo, Xiangyu Zhao·· 12 小时前AI 评分40
通过复杂任务语义分解实现自进化智能体工作流，获 GLEE 竞赛第一
Self-Evolving Agentic Workflows via Semantic Decomposition of Complex Tasks
AI 导读
研究者提出一种通过语义任务分解实现自进化的智能体工作流框架，每个子任务带有明确适用条件，并按需执行代码或调用模型，智能体利用执行反馈优化分解并更新流程。该工作在 GLEE 竞赛的议价、谈判与说服任务中测试，多数最终决策流程以代码实现，并获得该竞赛第一名。
正文 · 原文
Abstract:Agentic workflows organize complex tasks into reusable procedures. However, invoking an LLM at every step incurs repeated inference costs and delays. We propose a framework for self-evolving agentic workflows through semantic task decomposition. Each subtask has explicit applicability conditions and a procedure that executes code or invokes a model as needed. An agent uses execution feedback to refine the decomposition and update these procedures. Experiments in the GLEE competition show the framework on bargaining, negotiation and persuasion. Most observed final-decision procedures are implemented in code. Code is available at this https URL.
| Comments: | Accepted as a Spotlight at the NeurIPS 2026 Workshop on Interpreting Agent Behavior (IAB), GLEE Competition Track; 1st place in the GLEE competition. 21 pages |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Multiagent Systems (cs.MA) |
| Cite as: | arXiv:2611.01840 [cs.AI] |
| (or arXiv:2611.01840v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.01840 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zichuan Fu [view email]
[v1]
Sun, 1 Nov 2026 12:18:19 UTC (452 KB)
来源：arXiv cs.CL · arxiv.org