On the Robustness of Medical Vision-Language Models: Are They Truly Generalizable?
Imam, Raza Marew, Rufael Yaqub, Mohammad
Imam, Raza
Marew, Rufael
Yaqub, Mohammad
Supervisor
Department
Computer Vision
Embargo End Date
Type
Conference proceeding
Date
2026
License
Language
English
Collections
Research Projects
Organizational Units
Journal Issue
Abstract
Medical Vision-Language Models (MVLMs) have achieved par excellence generalization in medical image analysis, yet their performance under noisy, corrupted conditions remains largely untested. Clinical imaging is inherently susceptible to acquisition artifacts and noise; however, existing evaluations predominantly assess generally clean datasets, overlooking robustness—i.e., the model’s ability to perform under real-world distortions. To address this gap, we first introduce MediMeta-C, a corruption benchmark that systematically applies several perturbations across multiple medical imaging datasets. Combined with MedMNIST-C, this establishes a comprehensive robustness evaluation framework for MVLMs. We further propose RobustMedCLIP, a visual encoder adaptation of a pretrained MVLM that incorporates few-shot tuning to enhance resilience against corruptions. Through extensive experiments, we benchmark 5 major MVLMs across 5 medical imaging modalities, revealing that existing models exhibit severe degradation under corruption and struggle with domain-modality tradeoffs. Our findings highlight the necessity of diverse training and robust adaptation strategies, demonstrating that efficient low-rank adaptation when paired with few-shot tuning, improves robustness while preserving generalization across modalities.
Citation
R. Imam, R. Marew, and M. Yaqub, “On the Robustness of Medical Vision-Language Models: Are They Truly Generalizable?,” Lecture Notes in Computer Science, vol. 15916 LNCS, pp. 233–256, 2026, doi: 10.1007/978-3-031-98688-8_17/FIGURES/15
Source
Medical Image Understanding and Analysis
Conference
29th Annual Conference on Medical Image Understanding and Analysis, MIUA 2025
Keywords
Generalization, Healthcare, Medical VLM, Robustness
Subjects
Source
29th Annual Conference on Medical Image Understanding and Analysis, MIUA 2025
Publisher
Springer Nature