Reproduce “Reporting Practice Matters: The Impact of Reference Choice on Chest X-ray Report Evaluation” 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
Daniel P. Jeong, Charles Q. Li, Hossein Hosseiny, Nitya M. Bhalla, Fatma Uyar Morency, Pradeep Ravikumar, Zachary C. Lipton, Michael Oberst
ReRef tests how clinically equivalent chest X-ray references affect automated evaluation of Findings-section generation. A language model generator and verifier create reporting-style variations. On MIMIC-CXR under RadCliQ-v1, a minimal Findings reference moves Libra from first to second and CheXOne from third to first among nine models. This shows rankings can depend on reference choice, not clinical superiority or generalization beyond this task.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.