Reproduce “From Tool State to Scientific Action: Auditing Decision Provenance in Biomedical Research 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
Xingyu Yi, Xinya Zhang, Yunshen Zhou, Haoyu Bao
BioAgent-DF replays saved numerical tool outputs at original computation sites and measures evidence paths separately from verdicts. Across eight SciFact batches, seven selected combinations showed repeatable read or citation changes, while verdicts stayed unchanged and restoration recovered reference evidence. In hERG allocation, state restoration recovered all 39 changed decisions; findings concern selected cases and computational decisions, not biological truth.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.