Enhancing the Transferability of Adversarial Attacks through Variance Tuning [PDF]
Computer Vision and Pattern Recognition, 2021 Deep neural networks are vulnerable to adversarial examples that mislead the models with imperceptible perturbations. Though adversarial attacks have achieved incredible success rates in the white-box setting, most existing adversaries often exhibit weak
Xiaosen Wang, Kun He
semanticscholar +1 more source
Set-level Guidance Attack: Boosting Adversarial Transferability of Vision-Language Pre-training Models [PDF]
IEEE International Conference on Computer Vision, 2023 Vision-language pre-training (VLP) models have shown vulnerability to adversarial examples in multimodal tasks. Furthermore, malicious adversaries can be deliberately transferred to attack other black-box models. However, existing work has mainly focused
Dong Lu +5 more
semanticscholar +1 more source
Structure Invariant Transformation for better Adversarial Transferability [PDF]
IEEE International Conference on Computer Vision, 2023 Given the severe vulnerability of Deep Neural Networks (DNNs) against adversarial examples, there is an urgent need for an effective adversarial attack to identify the deficiencies of DNNs in security-sensitive applications. As one of the prevalent black-
Xiaosen Wang +2 more
semanticscholar +1 more source
Boosting Adversarial Transferability by Block Shuffle and Rotation [PDF]
Computer Vision and Pattern Recognition, 2023 Adversarial examples mislead deep neural networks with imperceptible perturbations and have brought significant threats to deep learning. An important aspect is their transferability, which refers to their ability to deceive other models, thus enabling ...
Kunyu Wang +3 more
semanticscholar +1 more source
Boosting Adversarial Transferability by Achieving Flat Local Maxima [PDF]
Neural Information Processing Systems, 2023 Transfer-based attack adopts the adversarial examples generated on the surrogate model to attack various models, making it applicable in the physical world and attracting increasing interest.
Zhijin Ge +4 more
semanticscholar +1 more source
Admix: Enhancing the Transferability of Adversarial Attacks [PDF]
IEEE International Conference on Computer Vision, 2021 Deep neural networks are known to be extremely vulnerable to adversarial examples under white-box setting. Moreover, the malicious adversaries crafted on the surrogate (source) model often exhibit black-box transferability on other models with the same ...
Xiaosen Wang +3 more
semanticscholar +1 more source
Improving Adversarial Transferability via Neuron Attribution-based Attacks [PDF]
Computer Vision and Pattern Recognition, 2022 Deep neural networks (DNNs) are known to be vulnerable to adversarial examples. It is thus imperative to devise effective attack algorithms to identify the deficiencies of DNNs beforehand in security-sensitive applications.
Jianping Zhang +6 more
semanticscholar +1 more source
Improving the Transferability of Adversarial Samples by Path-Augmented Method [PDF]
Computer Vision and Pattern Recognition, 2023 Deep neural networks have achieved unprecedented success on diverse vision tasks. However, they are vulnerable to adversarial noise that is imperceptible to humans.
Jianping Zhang +7 more
semanticscholar +1 more source
Boosting the Transferability of Adversarial Attacks with Reverse Adversarial Perturbation [PDF]
Neural Information Processing Systems, 2022 Deep neural networks (DNNs) have been shown to be vulnerable to adversarial examples, which can produce erroneous predictions by injecting imperceptible perturbations.
Zeyu Qin +6 more
semanticscholar +1 more source
Making Substitute Models More Bayesian Can Enhance Transferability of Adversarial Examples [PDF]
International Conference on Learning Representations, 2023 The transferability of adversarial examples across deep neural networks (DNNs) is the crux of many black-box attacks. Many prior efforts have been devoted to improving the transferability via increasing the diversity in inputs of some substitute models ...
Qizhang Li, Yiwen Guo, W. Zuo, Hao Chen
semanticscholar +1 more source