Attention Masks Help Adversarial Attacks to Bypass Safety Detectors

Yunfan Shi
11/7/2024
cs.CRcs.AI

No official code found. Best implementation: FrankShi9/Attention-Mask-Attack (1 star, last commit 2024-11-13).

Abstract

Despite recent research advancements in adversarial attack methods, current approaches against XAI monitors are still discoverable and slower. In this paper, we present an adaptive framework for attention mask generation to enable stealthy, explainable and efficient PGD image classification adversarial attack under XAI monitors. Specifically, we utilize mutation XAI mixture and multitask self-supervised X-UNet for attention mask generation to guide PGD attack. Experiments on MNIST (MLP), CIFAR-10 (AlexNet) have shown that our system can outperform benchmark PGD, Sparsefool and SOTA SINIFGSM in balancing among stealth, efficiency and explainability which is crucial for effectively fooling SOTA defense protected classifiers.

View on arXivView PDF

Code Implementations(1)

FrankShi9/Attention-Mask-AttackCommunity100%
10Python, Jupyter NotebookApr 2, 20231 years ago

Similar Papers

Self-Supervised Representation Learning for Adversarial Attack Detection

86% match2026

Cross-Task Attack: A Self-Supervision Generative Framework Based on Attention Shift

85% match2024

Towards Robust Vision Transformer via Masked Adaptive Ensemble

85% match2024

Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization

85% match2024

NumbOD: A Spatial-Frequency Fusion Attack Against Object Detectors

83% match1 repo82024

Unleashing the Power of Pre-trained Encoders for Universal Adversarial Attack Detection

83% match2025

Cite this paper

@article{shi2024attention,
  title  = {Attention Masks Help Adversarial Attacks to Bypass Safety Detectors},
  author = {Yunfan Shi},
  year   = {2024},
  eprint = {2411.04772},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2411.04772v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2411.04772.svg)](https://codeofpaper.com/paper/2411.04772)

Discussion