An Evaluation of Role-Based Multi-Agent Code Generation on Repository-Scale Problems

cs.SE arXiv:2607.04212
View PDF arXiv JSON

Abstract

Role-based multiagent code generation aims to make LLMs more effective on repository-scale problems, moving beyond small programming tasks. We evaluate this approach on 12 Java repositories, finding greater similarity to developer code than single LLMs, but a persistent gap from human implementations.

PDF Viewer