Metraipolis › archive › dc_01a8b406f9bb

Authority-Preserving Evaluation of Medical Vision-Language Assistants — arXiv:2609.22302

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.22302
License (as archived)
CC0-1.0
Archived
2026-10-09T21:59:09.168Z
SHA-256 of the archived text
36e5d99e5d2ab9a78003176e5e6a4955a7bbc342d736305da42625beefa47b9b
Stable URL of this page
https://metraipolis.com/documento/dc_01a8b406f9bb

How to cite

Authority-Preserving Evaluation of Medical Vision-Language Assistants — arXiv:2609.22302 — arXiv (en), https://arxiv.org/abs/2609.22302 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_01a8b406f9bb (Metraipolis archive dc_01a8b406f9bb, sha256 36e5d99e5d2ab9a78003176e5e6a4955a7bbc342d736305da42625beefa47b9b)
BibTeX
@misc{dc01a8b406f9bb,
  title = {Authority-Preserving Evaluation of Medical Vision-Language Assistants — arXiv:2609.22302},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.22302},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_01a8b406f9bb; sha256 36e5d99e5d2ab9a78003176e5e6a4955a7bbc342d736305da42625beefa47b9b},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22302. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Medical vision-language models can propose how urgently a skin lesion should be reviewed, but the local service retains authority to accept or replace that proposal under referral policy, capacity, and locally held patient context. Proposal quality and selected-action quality are therefore distinct evaluation targets, and benchmark evidence transfers between them only when local review preserves the expected action score. We introduce AuthEval, a logging and evaluation framework that records both actions, scores the selected action under declared local criteria, and, where feasible, scores the declined proposal under the same rule. It reports the resulting authority gap only when the record supports it. Because the gap is the product of the proposal-change rate and the mean score change on changed cases, that rate alone determines neither its magnitude nor its sign. On ISIC 2019, with MedGemma and simulated local review, two constraint regimes with similar change rates produced an optimistic image-equal gap under capacity ($+0.744$ simulator units) but no detectable gap under safety. The declared evaluation unit also mattered: under mixed constraints the gap reversed from $+0.374$ to $-0.206$ when weighting shifted from image to lesion-aware cluster. AuthEval thus clarifies whether a study's records support claims about the model, the workflow, or both.

Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.22302 — 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 2026-10-09T21:45:20.782Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).