VSF-Med:A Vulnerability Scoring Framework for Medical Vision-Language Models
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918076643016704 |
|---|---|
| author | Sadanandan, Binesh Behzadan, Vahid |
| author_facet | Sadanandan, Binesh Behzadan, Vahid |
| contents | Vision Language Models (VLMs) hold great promise for streamlining labour-intensive medical imaging workflows, yet systematic security evaluations in clinical settings remain scarce. We introduce VSF--Med, an end-to-end vulnerability-scoring framework for medical VLMs that unites three novel components: (i) a rich library of sophisticated text-prompt attack templates targeting emerging threat vectors; (ii) imperceptible visual perturbations calibrated by structural similarity (SSIM) thresholds to preserve clinical realism; and (iii) an eight-dimensional rubric evaluated by two independent judge LLMs, whose raw scores are consolidated via z-score normalization to yield a 0--32 composite risk metric. Built entirely on publicly available datasets and accompanied by open-source code, VSF--Med synthesizes over 30,000 adversarial variants from 5,000 radiology images and enables reproducible benchmarking of any medical VLM with a single command. Our consolidated analysis reports mean z-score shifts of $0.90σ$ for persistence-of-attack-effects, $0.74σ$ for prompt-injection effectiveness, and $0.63σ$ for safety-bypass success across state-of-the-art VLMs. Notably, Llama-3.2-11B-Vision-Instruct exhibits a peak vulnerability increase of $1.29σ$ for persistence-of-attack-effects, while GPT-4o shows increases of $0.69σ$ for that same vector and $0.28σ$ for prompt-injection attacks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_00052 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VSF-Med:A Vulnerability Scoring Framework for Medical Vision-Language Models Sadanandan, Binesh Behzadan, Vahid Computer Vision and Pattern Recognition Artificial Intelligence Vision Language Models (VLMs) hold great promise for streamlining labour-intensive medical imaging workflows, yet systematic security evaluations in clinical settings remain scarce. We introduce VSF--Med, an end-to-end vulnerability-scoring framework for medical VLMs that unites three novel components: (i) a rich library of sophisticated text-prompt attack templates targeting emerging threat vectors; (ii) imperceptible visual perturbations calibrated by structural similarity (SSIM) thresholds to preserve clinical realism; and (iii) an eight-dimensional rubric evaluated by two independent judge LLMs, whose raw scores are consolidated via z-score normalization to yield a 0--32 composite risk metric. Built entirely on publicly available datasets and accompanied by open-source code, VSF--Med synthesizes over 30,000 adversarial variants from 5,000 radiology images and enables reproducible benchmarking of any medical VLM with a single command. Our consolidated analysis reports mean z-score shifts of $0.90σ$ for persistence-of-attack-effects, $0.74σ$ for prompt-injection effectiveness, and $0.63σ$ for safety-bypass success across state-of-the-art VLMs. Notably, Llama-3.2-11B-Vision-Instruct exhibits a peak vulnerability increase of $1.29σ$ for persistence-of-attack-effects, while GPT-4o shows increases of $0.69σ$ for that same vector and $0.28σ$ for prompt-injection attacks. |
| title | VSF-Med:A Vulnerability Scoring Framework for Medical Vision-Language Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2507.00052 |