Imperceptible adversarial attacks aim to fool DNNs by adding imperceptible perturbation to the input data.AdvAD is a novel non-parametric diffusion-based method for crafting imperceptible adversarial examples.AdvAD achieves high attack efficacy and imperceptibility with lower overall perturbation strength.Extensive experiments demonstrate the effectiveness of AdvAD, outperforming state-of-the-art imperceptible attacks.