Reproduce “Multi-Agent LLM Collaboration for Unit Test Generation via Human-Testing-Inspired Workflows” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Quanjun Zhang, Ye Shang, Siqi Gu, Jianyi Zhou, Chunrong Fang, Zhenyu Chen, Liang Xiao
TestAgent generates unit tests through separate planning, generation, and review roles, using repository context and execution feedback to revise its work. On six Java projects with 1,454 focal methods, it reports 92.34% line coverage, 90.24% branch coverage, and 83.69% mutation score. Removing its knowledge graph lowers these measures; results come from a single benchmark run.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.