# Uncertainty Is Not a Safety Net for Clinical VQA, but Can It Anticipate Model Failure?

Source: [Uncertainty Is Not a Safety Net for Clinical VQA, but Can It Anticipate Model Failure?](<https://www.sota2.com/research/paper/6a30d8e3d33dfb23e783d011>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-06-17T05:00:21.724Z

Authors: Arnisa Fazla, Alberto Testoni, Ameen Abu-Hanna, Barbara Plank, Iacer Calixto

Published: 2026-06-15T11:27:11.000Z

## Abstract

Safe deployment of clinical vision-language models (VLMs) requires reliable uncertainty estimation (UE): a signal indicating when predictions should be trusted or escalated to a clinician. We test whether current UE methods actually deliver this signal. Benchmarking 8 methods across 12 VLMs on clinical visual question-answering (VQA), we find that UE quality is not an intrinsic property of the UE method: it tracks model accuracy, degrading precisely where the model performance is weakest, and therefore where reliability is most needed. When we stress-test models by hiding the correct option among the multiple-choice answers (NOTA perturbations), accuracy collapses while uncertainty barely changes, leaving models systematically miscalibrated. Yet, we find that uncertainty on the unperturbed input reliably anticipates which predictions will collapse under NOTA, indicating that UE in current VLMs carries diagnostic information about model fragility. Our results position UE as a diagnostic tool for identifying fragile predictions and motivate perturbation-based evaluation as a path toward safe clinical deployment.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Clinical Visual Question Answering on GMAI-MMBench standard MCQA](<https://www.sota2.com/research/sota/clinical-visual-question-answering-on-gmai-mmbench-standard-mcqa.md>): Clinical Visual Question Answering; GMAI-MMBench standard MCQA; Accuracy: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.16583>)

[PDF](<https://arxiv.org/pdf/2606.16583.pdf>)