Self-Evolving Agentic Workflows via Semantic Decomposition of Complex Tasks
Zichuan Fu, Junjia Qi, Xian Wu, Wenlin Zhang, Yimin Deng, Kaifeng Guo, Xiangyu Zhao
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 https://github.com/Applied-Machine-Learning-Lab/NeurIPS2026SeaFlow.
Create a lesson
Related papers
OSGR: Observation-Support Graph Reasoning for Earth Observation Question Answering
Dazhou Yu, Zerong Wang, Riyang Bao et al.
Bidirectional Centered Score Matching for Discrete Diffusion Distillation
Peng Tu, Baoyou Chen, Haoyuan Xia et al.
Harness-S: Learning from Sibling Harnesses for Generalizable Agent Training
Kangning Zhang, Wenxiang Jiao, Bowen Fang et al.
Reliability Stress Testing of RAG Retrieval under Query-Relevant Corpus Poisoning
Janet Sung, Kainan Zhou, Gangzhen Qian et al.
Neologism Inoculation Prompting
David Demitri Africa, Cameron Holmes
AceAction-5M: Hierarchical Goals for Generalist Action Learning Across 3D Game Worlds
Kailing Li, Ruoxuan Zhang, Zheyuan Zhou et al.