Metraipolis › archive › dc_c67abade05d2

Reliability Stress Tests and Decision-Time Routing for Chest X-ray Vision-Language Models — arXiv:2610.02270

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.02270
License (as archived)
CC0-1.0
Archived
2026-10-06T09:09:17.111Z
SHA-256 of the archived text
e7af8d9be1b90b84fa8b4c0d0d0852d5474543ac7967cd980873ea44ed34a1ad
Stable URL of this page
https://metraipolis.com/documento/dc_c67abade05d2

How to cite

Reliability Stress Tests and Decision-Time Routing for Chest X-ray Vision-Language Models — arXiv:2610.02270 — arXiv (en), https://arxiv.org/abs/2610.02270 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c67abade05d2 (Metraipolis archive dc_c67abade05d2, sha256 e7af8d9be1b90b84fa8b4c0d0d0852d5474543ac7967cd980873ea44ed34a1ad)
BibTeX
@misc{dcc67abade05d2,
  title = {Reliability Stress Tests and Decision-Time Routing for Chest X-ray Vision-Language Models — arXiv:2610.02270},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.02270},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c67abade05d2; sha256 e7af8d9be1b90b84fa8b4c0d0d0852d5474543ac7967cd980873ea44ed34a1ad},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.02270. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Medical vision-language model (VLM) evaluation is sensitive to workflow design, prompting strategy, and benchmark construction, yet most studies treat these factors in isolation. We introduce a reliability stress test for chest X-ray interpretation built on two balanced datasets (a private report-backed set and a curated MIMIC subset). Three medical VLMs (CheXagent, MedGemma-4B, and MedGemma-27B) are evaluated across three prompt styles and two workflows (single-VLM and multi-agent), producing 36 configurations. We show that exact-match accuracy alone can overstate the effectiveness of conservative models that default to "Normal" predictions. Diagnostic reliability also depends heavily on model family and scale: multi-agent reasoning helps some configurations but hurts others. Building on these observations, we propose a decision-time routing framework that selectively escalates to multi-agent inference only when beneficial, improving the cost-quality trade-off over fixed workflows. Our results highlight the need for evaluation protocols that jointly consider prompt sensitivity, failure-mode diversity, and workflow choice before clinical deployment.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-10-05 in cs.CV).
Categorie arXiv: Image and Video Processing (eess.IV) ; Computer Vision and Pattern Recognition (cs.CV).
Pagina del lavoro: https://arxiv.org/abs/2610.02270 — 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 2026-10-06T08:21:44.286Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-10-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).