Reproduce “Collaborative Agents for Automated Program Repair in Ruby” 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
Nikta Akbarpour, Mahdieh Sadat Benis, Fatemeh Hendijani Fard, Ali Ouni, Mohamed Aymen Saied
RAMP addresses Ruby automated program repair by combining agent-generated tests, bug hypotheses, and execution feedback to guide repair attempts, without a cross-language pipeline. On a 34-question sample of XCodeEval Ruby validation tasks, it achieves 67.0% cumulative pass@1, compared with LANTERN’s 61.7%. This benchmark result does not establish performance on production Ruby projects, and component effects vary by model.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.