A survey on transferability of adversarial examples across deep neural networks

J Gu, X Jia, P de Jorge, W Yu, X Liu, A Ma… - arXiv preprint arXiv …, 2023 - arxiv.org
The emergence of Deep Neural Networks (DNNs) has revolutionized various domains,
enabling the resolution of complex tasks spanning image recognition, natural language …

Comparative evaluation of recent universal adversarial perturbations in image classification

J Weng, Z Luo, D Lin, S Li - Computers & Security, 2024 - Elsevier
Abstract The vulnerability of Convolutional Neural Networks (CNNs) to adversarial samples
has recently garnered significant attention in the machine learning community. Furthermore …

Learning transferable targeted universal adversarial perturbations by sequential meta-learning

J Weng, Z Luo, D Lin, S Li - Computers & Security, 2024 - Elsevier
Recently, the transferability of adversarial perturbations in non-targeted scenarios has been
extensively studied. However, changing the predictions of an unknown model to a pre …

LimeAttack: Local Explainable Method for Textual Hard-Label Adversarial Attack

H Zhu, Q Zhao, W Shang, Y Wu, K Liu - Proceedings of the AAAI …, 2024 - ojs.aaai.org
Natural language processing models are vulnerable to adversarial examples. Previous
textual adversarial attacks adopt model internal information (gradients or confidence scores) …

Beamattack: Generating high-quality textual adversarial examples through beam search and mixed semantic spaces

H Zhu, Q Zhao, Y Wu - Pacific-Asia Conference on Knowledge Discovery …, 2023 - Springer
Natural language processing models based on neural networks are vulnerable to
adversarial examples. These adversarial examples are imperceptible to human readers but …

Non-cooperative game theory with generative adversarial network for effective decision-making in military cyber warfare

X Ma, W Abdelfattah, D Luo, N Innab… - Annals of Operations …, 2024 - Springer
Cyber warfare has become a critical domain of modern military operations, characterized by
the constant interplay of offense and defense in the digital realm. This research explores a …

BufferSearch: Generating Black-Box Adversarial Texts With Lower Queries

W Lv, Z Wang, Y Zheng, Z Zhong, Q Xuan… - arXiv preprint arXiv …, 2023 - arxiv.org
Machine learning security has recently become a prominent topic in the natural language
processing (NLP) area. The existing black-box adversarial attack suffers prohibitively from …

MAT: mixed-strategy game of adversarial training in fine-tuning

Z Zhong, T Chen, Z Wang - arXiv preprint arXiv:2306.15826, 2023 - arxiv.org
Fine-tuning large-scale pre-trained language models has been demonstrated effective for
various natural language processing (NLP) tasks. Previous studies have established that …

DeepMC: DNN test sample optimization method jointly guided by misclassification and coverage

J Sun, J Li, S Wen - Applied Intelligence, 2023 - Springer
Large-scale and high-quality test samples are extremely scarce in deep neural networks
(DNN) testing. Existing test sample optimization methods exhibit the problem of low …

A Framework to Enhance Security and Safety of Deep Learning Models Against Out-of-Distribution Examples

A Azmoodeh - 2024 - atrium.lib.uoguelph.ca
In recent years, the realm of deep learning has witnessed remarkable progress, with models
rooted in this paradigm surpassing traditional algorithms and, in some instances, human …