Enhancing Adversarial Transferability via Fourier-Based Input Transformation
Big Data and Cognitive Computing · 27.04.2026 · 10 (5) · 135
DOI: 10.3390/bdcc10050135 ↗Kurz gefasst
Fachartikel in der Zeitschrift Big Data and Cognitive Computing, online erschienen am 27.04.2026. Verfasst von Zilin Tian und 3 weiteren Autor:innen. Die Arbeit nennt 57 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Adversarial transferability makes black-box attacks practical and exposes weaknesses of deep neural networks for computer vision, image recognition, and visual understanding. Among various transferability-enhancing methods, input transformation is one of the most effective strategies. However, existing methods often ignore the decoupling of style and semantics in the input image, as well as the need for customized transformation strategies, resulting in limited performance gains or suboptimal outcomes. In this paper, we propose a novel Fourier-based perspective for input transformation generalization in the context of vision adversarial attacks. The main observations are that the Fourier amplitude captures stylistic information and the phase encompasses richer semantics which are crucial for visual understanding. Motivated by this, we develop a Fourier-based strategy, which performs a stylistic transform and semantic mixup on the input examples to improve transferability. To avoid inconsistent semantics of augmented images for the surrogate model, we mix the original images with the augmentations to maintain semantic consistency and mitigate imprecise gradients. Extensive experiments on ImageNet-compatible datasets demonstrate that our method consistently outperforms existing input transformation attacks.
Zitieren
APA
Tian, Z., Wang, X., Long, Y., & Zhang, L. (2026). Enhancing Adversarial Transferability via Fourier-Based Input Transformation. Big Data and Cognitive Computing, 10(5), 135. https://doi.org/10.3390/bdcc10050135
Vancouver
Tian Z, Wang X, Long Y, Zhang L. Enhancing Adversarial Transferability via Fourier-Based Input Transformation. Big Data and Cognitive Computing. 2026;10(5):135. doi:10.3390/bdcc10050135
BibTeX
@article{Tian2026,
author = {Tian, Zilin and Wang, Xin and Long, Yunfei and Zhang, Liguo},
title = {Enhancing Adversarial Transferability via Fourier-Based Input Transformation},
journal = {Big Data and Cognitive Computing},
year = {2026},
volume = {10},
number = {5},
pages = {135},
doi = {10.3390/bdcc10050135}
}
EUVIMED