CT-GAT: Cross-Task Generative Adversarial Attack based on Transferability
This addresses the risk of adversarial attacks in real-world scenarios where substitute models are impractical, offering a more efficient approach.
The paper tackles the problem of adversarial attacks by proposing a method that directly constructs adversarial examples using transferable features across tasks, achieving superior attack performance on ten datasets with small cost.
Neural network models are vulnerable to adversarial examples, and adversarial transferability further increases the risk of adversarial attacks. Current methods based on transferability often rely on substitute models, which can be impractical and costly in real-world scenarios due to the unavailability of training data and the victim model's structural details. In this paper, we propose a novel approach that directly constructs adversarial examples by extracting transferable features across various tasks. Our key insight is that adversarial transferability can extend across different tasks. Specifically, we train a sequence-to-sequence generative model named CT-GAT using adversarial sample data collected from multiple tasks to acquire universal adversarial features and generate adversarial examples for different tasks. We conduct experiments on ten distinct datasets, and the results demonstrate that our method achieves superior attack performance with small cost.