BIOENGINEER

AI Reviewers Judge the Author, Not Just the Paper, Study Finds

A controlled counterfactual audit of 80 arXiv manuscripts shows that large language model evaluators systematically inflate ...