Pith. sign in

REVIEW 6 cited by

CARES: A Comprehensive Benchmark of Trustworthiness in Medical Vision Language Models

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2406.06007 v3 pith:J2JKZ5GI submitted 2024-06-10 cs.LG cs.CLcs.CVcs.CY

classification cs.LGcs.CLcs.CVcs.CY
keywords medicaltrustworthinessmed-lvlmsacrosscaresmodelsbenchmarkfairness
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Artificial intelligence has significantly impacted medical applications, particularly with the advent of Medical Large Vision Language Models (Med-LVLMs), sparking optimism for the future of automated and personalized healthcare. However, the trustworthiness of Med-LVLMs remains unverified, posing significant risks for future model deployment. In this paper, we introduce CARES and aim to comprehensively evaluate the Trustworthiness of Med-LVLMs across the medical domain. We assess the trustworthiness of Med-LVLMs across five dimensions, including trustfulness, fairness, safety, privacy, and robustness. CARES comprises about 41K question-answer pairs in both closed and open-ended formats, covering 16 medical image modalities and 27 anatomical regions. Our analysis reveals that the models consistently exhibit concerns regarding trustworthiness, often displaying factual inaccuracies and failing to maintain fairness across different demographic groups. Furthermore, they are vulnerable to attacks and demonstrate a lack of privacy awareness. We publicly release our benchmark and code in https://cares-ai.github.io/.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 6 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Attention Without Grounding: Causal Evaluation of Visual Explanations in Medical VLMs

    cs.CV 2026-07 accept novelty 6.0 of 10

    Medical VLM attention and saliency heatmaps are not causally faithful: they miss radiologist-annotated regions and anti-correlate with patch-occlusion importance, unlike CXR classifier baselines.

  2. MedBLINK: Probing Basic Perception in Multimodal Language Models for Medicine

    cs.AI 2025-08 conditional novelty 6.0 of 10

    Current medical multimodal models, including GPT-4o and Claude 3.5 Sonnet, fail simple perceptual tasks on medical images that human experts solve almost perfectly.

  3. PAC Bench: Do Foundation Models Understand Prerequisites for Executing Manipulation Policies?

    cs.RO 2025-06 conditional novelty 6.0 of 10

    PAC Bench evaluates vision-language models on properties, affordances, and constraints for manipulation, finding near-zero performance on physical constraints.

  4. Focus on What Matters: Enhancing Medical Vision-Language Models with Automatic Attention Alignment Tuning

    cs.CV 2025-05 conditional novelty 6.0 of 10

    A3Tune aligns the visual attention of medical LVLMs to prompt-relevant regions via SAM and BioMedCLIP weak labels plus a Mixture-of-Experts over LoRA, improving VQA and report generation accuracy.

  5. The Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to Autonomy

    cs.AI 2026-07 conditional novelty 4.5 of 10

    Medical agents should be scaled mainly by richer clinical environments and self-evolution loops, not parameter growth alone, under a three-level autonomy taxonomy.

  6. Uncertainty-Driven Expert Control: Enhancing the Reliability of Medical Vision-Language Models

    cs.CV 2025-07 reject novelty 4.0 of 10

    Expert-CFG combines entropy-based uncertainty selection with classifier-free guidance over expert-highlighted text to refine MedVLM outputs, reporting gains on VQA-RAD, SLAKE, and PathVQA.

Pith tools