Open Access

ARTICLE

AMA: Adaptive Multimodal Adversarial Attack with Dynamic Perturbation Optimization

Yufei Shi, Ziwen He^*, Teng Jin, Haochen Tong, Zhangjie Fu

Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science and Technology, Nanjing, 210044, China

* Corresponding Author: Ziwen He. Email:

Computer Modeling in Engineering & Sciences 2025, 144(2), 1831-1848. https://doi.org/10.32604/cmes.2025.067658

Received 09 May 2025; Accepted 23 July 2025; Issue published 31 August 2025

Abstract

This article proposes an innovative adversarial attack method, AMA (Adaptive Multimodal Attack), which introduces an adaptive feedback mechanism by dynamically adjusting the perturbation strength. Specifically, AMA adjusts perturbation amplitude based on task complexity and optimizes the perturbation direction based on the gradient direction in real time to enhance attack efficiency. Experimental results demonstrate that AMA elevates attack success rates from approximately 78.95% to 89.56% on visual question answering and from 78.82% to 84.96% on visual reasoning tasks across representative vision-language benchmarks. These findings demonstrate AMA’s superior attack efficiency and reveal the vulnerability of current visual language models to carefully crafted adversarial examples, underscoring the need to enhance their robustness.

---

Keywords

---