Health AI Central
Share:
Peer-Reviewed Publication
Pac Symp Biocomput2026;31251-264.January 1, 2026Journal Article

ReXVQA: A Large-scale Visual Question Answering Benchmark for Generalist Chest X-ray Understanding.

Ankit Pal1, Jung-Oh Lee2, Xiaoman Zhang3, Malaikannan Sankarasubbu4, Seunghyeon Roh5, Won Jung Kim6, Meesun Lee7, Pranav Rajpurkar8
1Saama AI Research, Saama Technologies, India3Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA, ankit.pal@saama.com.
2Seoul National University, Seoul, South Korea, pisceanoh@snu.ac.kr.
3Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA, xiaoman.zhang@hms.harvard.edu.
4Saama AI Research, Saama Technologies, India, malaikannan.sankarasubbu@saama.com.
5Seoul National University, Seoul, South Korea, seunghyeon.roh@snu.ac.kr.
6Seoul National University, Seoul, South Korea, wonjung.kim@snu.ac.kr.
7Seoul National University, Seoul, South Korea, meesun.lee@snu.ac.kr.
8Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA, pranav.rajpurkar@hms.harvard.edu.
PubMed DOI

Abstract

We present ReXVQA, the largest and most comprehensive benchmark for visual question answering (VQA) in chest radiology, comprising 694,841 questions paired with 160,000 chest X-rays studies across training, validation, and test sets. Unlike prior efforts that rely heavily on template based queries, ReXVQA introduces a diverse and clinically authentic task suite reflecting five core radiological re…

Create a free account to keep reading

Free members get 10 full research views every month across publications, clinical trials, FDA clearances, adverse events, and NIH grants. No credit card required.

Create Free AccountSign In

Want unlimited research access? See Pro plans

Data Accuracy Notice: Research intelligence on Health AI Central is aggregated from public sources (PubMed, ClinicalTrials.gov, FDA, NIH, CMS, and others) and refreshed nightly. Classifications and derived metrics are produced by automated methods described in our Methodology. We recommend verifying critical data points against the primary sources before making decisions.