REAgent: Requirement-Driven LLM Agents for Software Issue Resolution

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kuang, Shiqi, Tian, Zhao, Lin, Kaiwei, Tao, Chaofan, Wang, Shaowei, Bai, Haoli, Shang, Lifeng, Chen, Junjie
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908947458293760
author Kuang, Shiqi
Tian, Zhao
Lin, Kaiwei
Tao, Chaofan
Wang, Shaowei
Bai, Haoli
Shang, Lifeng
Chen, Junjie
author_facet Kuang, Shiqi
Tian, Zhao
Lin, Kaiwei
Tao, Chaofan
Wang, Shaowei
Bai, Haoli
Shang, Lifeng
Chen, Junjie
contents Issue resolution aims to automatically generate patches from given issue descriptions and has attracted significant attention with the rapid advancement of large language models (LLMs). However, due to the complexity of software issues and codebases, LLM-generated patches often fail to resolve corresponding issues. Although various advanced techniques have been proposed with carefully designed tools and workflows, they typically treat issue descriptions as direct inputs and largely overlook their quality (e.g., missing critical context or containing ambiguous information), which hinders LLMs from accurate understanding and resolution. To address this limitation, we draw on principles from software requirements engineering and propose REAgent, a requirement-driven LLM agent framework that introduces issue-oriented requirements as structured task specifications to better guide patch generation. Specifically, REAgent automatically constructs structured and information-rich issue-oriented requirements, identifies low-quality requirements, and iteratively refines them to improve patch correctness. We conduct comprehensive experiments on three widely used benchmarks using two advanced LLMs, comparing against five representative or state-of-the-art baselines. The results demonstrate that REAgent consistently outperforms all baselines, achieving an average improvement of 17.40% in terms of the number of successfully-resolved issues (% Resolved).
format Preprint
id arxiv_https___arxiv_org_abs_2604_06861
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle REAgent: Requirement-Driven LLM Agents for Software Issue Resolution
Kuang, Shiqi
Tian, Zhao
Lin, Kaiwei
Tao, Chaofan
Wang, Shaowei
Bai, Haoli
Shang, Lifeng
Chen, Junjie
Software Engineering
Issue resolution aims to automatically generate patches from given issue descriptions and has attracted significant attention with the rapid advancement of large language models (LLMs). However, due to the complexity of software issues and codebases, LLM-generated patches often fail to resolve corresponding issues. Although various advanced techniques have been proposed with carefully designed tools and workflows, they typically treat issue descriptions as direct inputs and largely overlook their quality (e.g., missing critical context or containing ambiguous information), which hinders LLMs from accurate understanding and resolution. To address this limitation, we draw on principles from software requirements engineering and propose REAgent, a requirement-driven LLM agent framework that introduces issue-oriented requirements as structured task specifications to better guide patch generation. Specifically, REAgent automatically constructs structured and information-rich issue-oriented requirements, identifies low-quality requirements, and iteratively refines them to improve patch correctness. We conduct comprehensive experiments on three widely used benchmarks using two advanced LLMs, comparing against five representative or state-of-the-art baselines. The results demonstrate that REAgent consistently outperforms all baselines, achieving an average improvement of 17.40% in terms of the number of successfully-resolved issues (% Resolved).
title REAgent: Requirement-Driven LLM Agents for Software Issue Resolution
topic Software Engineering
url https://arxiv.org/abs/2604.06861