Metraipolis › archive › dc_7db79be5946d

Med-AR: Autoregressive Vision-Language Pretraining for Long-Tailed Chest X-Ray Classification and Uncertainty-Aware Evaluation — arXiv:2609.29156

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.29156
License (as archived)
CC0-1.0
Archived
2026-10-08T04:27:19.750Z
SHA-256 of the archived text
d3f16921abbd7cc6e4f24fe00790e2f44901c718f552f61272bfa550d22142b4
Stable URL of this page
https://metraipolis.com/documento/dc_7db79be5946d

How to cite

Med-AR: Autoregressive Vision-Language Pretraining for Long-Tailed Chest X-Ray Classification and Uncertainty-Aware Evaluation — arXiv:2609.29156 — arXiv (en), https://arxiv.org/abs/2609.29156 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7db79be5946d (Metraipolis archive dc_7db79be5946d, sha256 d3f16921abbd7cc6e4f24fe00790e2f44901c718f552f61272bfa550d22142b4)
BibTeX
@misc{dc7db79be5946d,
  title = {Med-AR: Autoregressive Vision-Language Pretraining for Long-Tailed Chest X-Ray Classification and Uncertainty-Aware Evaluation — arXiv:2609.29156},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.29156},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7db79be5946d; sha256 d3f16921abbd7cc6e4f24fe00790e2f44901c718f552f61272bfa550d22142b4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.29156. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Long-tailed chest X-ray classification requires visual representations that capture both common abnormalities and subtle, infrequent findings. We propose Med-AR-8B and Med-AR-2B, two radiology-native autoregressive vision-language models pretrained with structured reports, abnormality-focused text, and region annotations. We evaluate the transfer of their visual encoders to multi-label classification against contrastive, self-supervised, and supervised pretrained encoders, including Med-CLIP, CheXFound, EVA-Base, ARK, and BioViL-T, using a common ML-Decoder classification head. To assess fine-grained recognition, we also construct LLM-expanded, report-derived label sets for MIMIC-CXR and CheXpert. Across PadChest, MIMIC-CXR, and CheXpert, Med-AR-8B outperforms Med-CLIP in mean AUROC and AUPRC for head, medium, and tail findings. On MIMIC-CXR, it increases tail-label mean AUPRC from 0.1033 to 0.1441. Med-AR-2B achieves the strongest discrimination results on PadChest. Across the broader encoder comparison, a Med-AR variant achieves the highest mean AUROC and AUPRC in every reported prevalence group on each public dataset. Both Med-AR variants also achieve lower excess area under the risk-coverage curve than Med-CLIP on all three public datasets, indicating improved selective-prediction performance under the evaluated protocol. Internal results are metric-dependent, with Med-CLIP retaining advantages in overall and tail AUPRC and in selective prediction. These findings establish Med-AR as a strong pretraining recipe for long-tailed chest X-ray classification on the evaluated public benchmarks and demonstrate the value of assessing discrimination and selective prediction together.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-25 in cs.AI).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.29156 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T03:50:58.015Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-25?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).