Type: url
Original: https://arxiv.org/pdf/2404.04834
Status: generated
Summary
This paper systematically reviews 71 primary studies on LLM-based multi-agent (LMA) systems in software engineering, organizing them across requirements engineering, code generation, quality assurance, maintenance, and end-to-end development. It defines an LMA system as an orchestration platform together with LLM-based agents, and reports two case studies in which ChatDev is used to build Snake and Tetris games. It closes with a two-phase research agenda covering individual agent capabilities and agent synergy.