# MedVIGIL: Evaluating Trustworthy Medical VLMs Under Broken Visual Evidence

arXiv: https://arxiv.org/abs/2605.07919 · HTML: https://mlsanity.com/paper/2605.07919/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 34 of its 39 references that are in the map.
- First posted on arXiv 8 May 2026.

## Builds on (most cited first, 30 of 34)

- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Making the V in VQA Matter: Elevating the Role of Image Understanding in Visual Question Answering](https://mlsanity.com/paper/1612.00837/) (2016)
- [Large Language Models Encode Clinical Knowledge](https://mlsanity.com/paper/2212.13138/) (2022)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison](https://mlsanity.com/paper/1901.07031/) (2019)
- [Datasheets for Datasets](https://mlsanity.com/paper/1803.09010/) (2018)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Object Hallucination in Image Captioning](https://mlsanity.com/paper/1809.02156/) (2018)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [Med-Flamingo: a Multimodal Medical Few-shot Learner](https://mlsanity.com/paper/2307.15189/) (2023)
- [Domain Adaptation for Medical Image Analysis: A Survey](https://mlsanity.com/paper/2102.09508/) (2021)
- [PMC-VQA: Visual Instruction Tuning for Medical Visual Question Answering](https://mlsanity.com/paper/2305.10415/) (2023)
- [HuatuoGPT-Vision, Towards Injecting Medical Visual Knowledge into Multimodal LLMs at Scale](https://mlsanity.com/paper/2406.19280/) (2024)
- [Data Cards: Purposeful and Transparent Dataset Documentation for Responsible AI](https://mlsanity.com/paper/2204.01075/) (2022)
- [AbstentionBench: Reasoning LLMs Fail on Unanswerable Questions](https://mlsanity.com/paper/2506.09038/) (2025)
- [Know Your Limits: A Survey of Abstention in Large Language Models](https://mlsanity.com/paper/2407.18418/) (2024)
- [Towards Causal VQA: Revealing and Reducing Spurious Correlations by Invariant and Covariant Semantic Editing](https://mlsanity.com/paper/1912.07538/) (2019)
- [Cycle-Consistency for Robust Visual Question Answering](https://mlsanity.com/paper/1902.05660/) (2019)
- [Croissant: A Metadata Format for ML-Ready Datasets](https://mlsanity.com/paper/2403.19546/) (2024)
- [Self-Evaluation Improves Selective Generation in Large Language Models](https://mlsanity.com/paper/2312.09300/) (2023)
- [MIRAGE: The Illusion of Visual Understanding](https://mlsanity.com/paper/2603.21687/) (2026)
- [Detecting and Evaluating Medical Hallucinations in Large Vision Language Models](https://mlsanity.com/paper/2406.10185/) (2024)
- [BabyVision: Visual Reasoning Beyond Language](https://mlsanity.com/paper/2601.06521/) (2026)
- [VQA-LOL: Visual Question Answering under the Lens of Logic](https://mlsanity.com/paper/2002.08325/) (2020)
- [VLind-Bench: Measuring Language Priors in Large Vision-Language Models](https://mlsanity.com/paper/2406.08702/) (2024)
- [A Novel Evaluation Benchmark for Medical LLMs: Illuminating Safety and Effectiveness in Clinical Domains](https://mlsanity.com/paper/2507.23486/) (2025)
- [MedVH: Towards Systematic Evaluation of Hallucination for Large Vision Language Models in the Medical Context](https://mlsanity.com/paper/2407.02730/) (2024)
- [Localizing Before Answering: A Hallucination Evaluation Benchmark for Grounded Medical Multimodal LLMs](https://mlsanity.com/paper/2505.00744/) (2025)
- [MedHEval: Benchmarking Hallucinations and Mitigation Strategies in Medical Large Vision-Language Models](https://mlsanity.com/paper/2503.02157/) (2025)

## Abstract (arXiv)

Medical vision--language models (VLMs) are usually evaluated on intact image--question pairs, but trustworthy clinical use requires a stronger property: a model must recognise when the evidential basis for an answer has failed. We study this through silent failures under perturbed evidence, where a vision-required medical question is paired with a false premise, wording perturbation, knowledge-only rewrite, or ROI-corrupted image, yet the model returns a fluent non-refusal answer. We introduce medvigil, a 300-case evaluation suite drawn from four public medical VQA sources, supervised end to end by four board-certified radiologists: every gold answer, refusal option, candidate-answer set, paraphrase, false-premise trap, ROI box, and clinical risk tier is clinician-authored. Two attending radiologists annotate every case in parallel, a senior radiologist consolidates the released manifest, and a separate fourth radiologist independent of construction answers every probe to provide the human reference baseline. The release contains 2556 MCQ probes, 240 counterfactual triplets, physician-adjudicated risk-tier and answerability flags, ROI boxes, and a paired open-ended variant. We report seven correctness-conditioned audit metrics that summarise into the medvigil Composite Score (MCS), and audit 16 vision-capable models plus two text-only baselines. The independent radiologist scores MCS 83.3 at silent-failure rate 5.8%, leaving a 14.1-point composite headroom above the strongest audited model (Claude Opus 4.7 at 69.2). The benchmark and evaluation harness are publicly released.