Reproduce “BioXArena-Pro: a research-grade benchmark for autonomous biomedical machine learning agents” 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
Loka Li, Duzhen Zhang, Xingbo Du, Leonard Song, Kun Zhang, Le Song
BioXArena-Pro tests whether coding agents can build biomedical prediction pipelines that generalize to held-out biological or clinical groups. It uses 64 research-derived tasks with private test labels, group-held-out splits, integrity audits, and a two-hour, one-GPU budget. The best mean score is 55.1/100; agents tie published methods on nine close-to-original tasks and lead references on 48 rebuilt tasks, not human experts.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.