Diversity Empowers Intelligence: Integrating Expertise of Software Engineering Agents

  • 2024-08-13 18:50:28
  • Kexun Zhang, Weiran Yao, Zuxin Liu, Yihao Feng, Zhiwei Liu, Rithesh Murthy, Tian Lan, Lei Li, Renze Lou, Jiacheng Xu, Bo Pang, Yingbo Zhou, Shelby Heinecke, Silvio Savarese, Huan Wang, Caiming Xiong
  • 0

Abstract

Large language model (LLM) agents have shown great potential in solvingreal-world software engineering (SWE) problems. The most advanced open-sourceSWE agent can resolve over 27% of real GitHub issues in SWE-Bench Lite.However, these sophisticated agent frameworks exhibit varying strengths,excelling in certain tasks while underperforming in others. To fully harnessthe diversity of these agents, we propose DEI (Diversity EmpoweredIntelligence), a framework that leverages their unique expertise. DEI functionsas a meta-module atop existing SWE agent frameworks, managing agent collectivesfor enhanced problem-solving. Experimental results show that a DEI-guidedcommittee of agents is able to surpass the best individual agent's performanceby a large margin. For instance, a group of open-source SWE agents, with amaximum individual resolve rate of 27.3% on SWE-Bench Lite, can achieve a 34.3%resolve rate with DEI, making a 25% improvement and beating most closed-sourcesolutions. Our best-performing group excels with a 55% resolve rate, securingthe highest ranking on SWE-Bench Lite. Our findings contribute to the growingbody of research on collaborative AI systems and their potential to solvecomplex software engineering challenges.

 

Quick Read (beta)

loading the full paper ...