Language Guided Adversarial Purification. (arXiv:2309.10348v1 [cs.LG])

Adversarial purification using generative models demonstrates strong
adversarial defense performance. These methods are classifier and
attack-agnostic, making them versatile but often computationally intensive.
Recent strides in diffusion and score networks have improved image generation
and, by extension, adversarial purification. Another highly efficient class of
adversarial defense methods known as adversarial training requires specific
knowledge of attack vectors, forcing them to be trained extensively on
adversarial examples. To overcome these limitations, we introduce a new
framework, namely Language Guided Adversarial Purification …

adversarial attack class defense extension generative generative models image image generation knowledge language making networks performance score training