SOURCE-LINKED INTELLIGENCE

Authority-Preserving Evaluation of Medical Vision-Language Assistants

arXiv · AI, language, vision and robotics · article · Sep 14, 2026 · UTC

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

Read original source ↗ Open in workspace

recordType

paper

region

Global