SEW: Self-Evolving Agentic Workflows for Automated Code Generation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Siwei, Fang, Jinyuan, Zhou, Han, Wang, Yingxu, Meng, Zaiqiao
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908960790937600
author Liu, Siwei
Fang, Jinyuan
Zhou, Han
Wang, Yingxu
Meng, Zaiqiao
author_facet Liu, Siwei
Fang, Jinyuan
Zhou, Han
Wang, Yingxu
Meng, Zaiqiao
contents Large Language Models (LLMs) have demonstrated effectiveness in code generation tasks. To enable LLMs to address more complex coding challenges, existing research has focused on crafting multi-agent systems with agentic workflows, where complex coding tasks are decomposed into sub-tasks, assigned to specialized agents. Despite their effectiveness, current approaches heavily rely on hand-crafted agentic workflows, with both agent topologies and prompts manually designed, which limits their ability to automatically adapt to different types of coding problems. To address these limitations and enable automated workflow design, we propose \textbf{S}elf-\textbf{E}volving \textbf{W}orkflow (\textbf{SEW}), a novel self-evolving framework that automatically generates and optimises multi-agent workflows. Extensive experiments on three coding benchmark datasets, including the challenging LiveCodeBench, demonstrate that our SEW can automatically design agentic workflows and optimise them through self-evolution, bringing up to 12\% improvement on LiveCodeBench compared to using the backbone LLM only. Furthermore, by investigating different representation schemes of workflow, we provide insights into the optimal way to encode workflow information with text.
format Preprint
id arxiv_https___arxiv_org_abs_2505_18646
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SEW: Self-Evolving Agentic Workflows for Automated Code Generation
Liu, Siwei
Fang, Jinyuan
Zhou, Han
Wang, Yingxu
Meng, Zaiqiao
Software Engineering
Artificial Intelligence
Computation and Language
Large Language Models (LLMs) have demonstrated effectiveness in code generation tasks. To enable LLMs to address more complex coding challenges, existing research has focused on crafting multi-agent systems with agentic workflows, where complex coding tasks are decomposed into sub-tasks, assigned to specialized agents. Despite their effectiveness, current approaches heavily rely on hand-crafted agentic workflows, with both agent topologies and prompts manually designed, which limits their ability to automatically adapt to different types of coding problems. To address these limitations and enable automated workflow design, we propose \textbf{S}elf-\textbf{E}volving \textbf{W}orkflow (\textbf{SEW}), a novel self-evolving framework that automatically generates and optimises multi-agent workflows. Extensive experiments on three coding benchmark datasets, including the challenging LiveCodeBench, demonstrate that our SEW can automatically design agentic workflows and optimise them through self-evolution, bringing up to 12\% improvement on LiveCodeBench compared to using the backbone LLM only. Furthermore, by investigating different representation schemes of workflow, we provide insights into the optimal way to encode workflow information with text.
title SEW: Self-Evolving Agentic Workflows for Automated Code Generation
topic Software Engineering
Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2505.18646