Chest X-ray (CXR) interpretation is hindered by the long-tailed distribution of pathologies and the open-world nature of clinical environments. Existing benchmarks often rely on closed-set classes from single institutions, failing to capture the prevalence of rare diseases or the appearance of novel findings. To address this, we present the CXR-LT 2026 challenge. This third iteration of the benchmark introduces a multi-center dataset comprising over 145,000 images from PadChest and NIH Chest X-ray datasets. The challenge defines two core tasks: (1) Robust Multi-Label Classification on 30 known classes and (2) Open-World Generalization to 6 unseen (out-of-distribution) rare disease classes. We report the results of the top-performing teams, evaluating them via mean Average Precision (mAP), AUROC, and F1-score. The winning solutions achieved an mAP of 0.5854 on Task 1 and 0.4315 on Task 2, demonstrating that large-scale vision-language pre-training significantly mitigates the performance drop typically associated with zero-shot diagnosis.
@article{arxiv.2602.22092,
title = {Overview of the CXR-LT 2026 Challenge: Multi-Center Long-Tailed and Zero Shot Chest X-ray Classification},
author = {Hexin Dong and Yi Lin and Pengyu Zhou and Xuan Zhong Feng and Alan Clint Legasto and Mingquan Lin and Hao Chen and Yuzhe Yang and George Shih and Yifan Peng},
journal= {arXiv preprint arXiv:2602.22092},
year = {2026}
}