Saved in:
Bibliographic Details
Main Authors: Rashid, Maisha Binte, Rivas, Pablo
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.16361
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910843804844032
author Rashid, Maisha Binte
Rivas, Pablo
author_facet Rashid, Maisha Binte
Rivas, Pablo
contents Vision-Language Models (VLMs) are increasingly deployed in public sector missions, necessitating robust evaluation of their safety and vulnerability to adversarial attacks. This paper introduces a novel framework to quantify adversarial risks in VLMs. We analyze model performance under Gaussian, salt-and-pepper, and uniform noise, identifying misclassification thresholds and deriving composite noise patches and saliency patterns that highlight vulnerable regions. These patterns are compared against the Fast Gradient Sign Method (FGSM) to assess their adversarial effectiveness. We propose a new Vulnerability Score that combines the impact of random noise and adversarial attacks, providing a comprehensive metric for evaluating model robustness.
format Preprint
id arxiv_https___arxiv_org_abs_2502_16361
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle A Framework for Evaluating Vision-Language Model Safety: Building Trust in AI for Public Sector Applications
Rashid, Maisha Binte
Rivas, Pablo
Computers and Society
Computation and Language
Computer Vision and Pattern Recognition
I.2.10; I.4.9; K.4.1
Vision-Language Models (VLMs) are increasingly deployed in public sector missions, necessitating robust evaluation of their safety and vulnerability to adversarial attacks. This paper introduces a novel framework to quantify adversarial risks in VLMs. We analyze model performance under Gaussian, salt-and-pepper, and uniform noise, identifying misclassification thresholds and deriving composite noise patches and saliency patterns that highlight vulnerable regions. These patterns are compared against the Fast Gradient Sign Method (FGSM) to assess their adversarial effectiveness. We propose a new Vulnerability Score that combines the impact of random noise and adversarial attacks, providing a comprehensive metric for evaluating model robustness.
title A Framework for Evaluating Vision-Language Model Safety: Building Trust in AI for Public Sector Applications
topic Computers and Society
Computation and Language
Computer Vision and Pattern Recognition
I.2.10; I.4.9; K.4.1
url https://arxiv.org/abs/2502.16361