Transfer Based Attack

Transfer-based attacks exploit the vulnerability of deep neural networks (DNNs), particularly vision transformers (ViTs) and convolutional neural networks (CNNs), by crafting adversarial examples on a surrogate model and transferring them to attack unseen target models. Current research focuses on improving the transferability of these attacks across different model architectures, quantization levels, and datasets, employing techniques like input transformations, Bayesian formulations, and backpropagation path manipulation to enhance their effectiveness. This research is crucial for evaluating and improving the robustness of DNNs in various applications, ranging from image classification to speech recognition, ultimately contributing to the development of more secure and reliable AI systems.

Papers

August 3, 2024

Downstream Transfer Attack: Adversarial Attacks on Downstream Models with Pre-trained Vision Transformers
Weijie Zheng, Xingjun Ma, Hanxun Huang, Zuxuan Wu, Yu-Gang Jiang
Adversarial Attack Adversarial Example Pre Trained Vision Transformer Adversarial Vulnerability Pre Trained ViT Downstream Model Transfer Based Attack

May 15, 2024

Properties that allow or prohibit transferability of adversarial attacks among quantized networks
Abhishek Shrestha, Jürgen Großmann
Adversarial Attack Adversarial Example Network Programming Black Box Task Transferability Many Property Network Compression Transfer Based Attack

November 2, 2023

Towards Evaluating Transfer-based Attacks Systematically, Practically, and Fairly
Qizhang Li, Yiwen Guo, Wangmeng Zuo, Hao Chen
Adversarial Example Black Box New Attack Adversarial Vulnerability Transfer Based Attack Caution Aware Transfer

September 26, 2023

Structure Invariant Transformation for better Adversarial Transferability
Xiaosen Wang, Zeliang Zhang, Jianping Zhang
Adversarial Attack Adversarial Transferability Structural Attack Input Transformation Transfer Based Attack Effective Adversarial

August 20, 2023

Boosting Adversarial Transferability by Block Shuffle and Rotation
Kunyu Wang, Xuanran He, Wenxuan Wang, Xiaosen Wang
Adversarial Example Rotation Angle Shuffle Model Adversarial Transferability Input Transformation Transfer Based Attack

August 15, 2023

Backpropagation Path Search On Adversarial Transferability
Zhuoer Xu, Zhangxuan Gu, Jianping Zhang, Shiwen Cui, Changhua Meng, Weiqiang Wang
Adversarial Example Structural Attack Transfer Based Attack

July 21, 2023

Improving Transferability of Adversarial Examples via Bayesian Attacks
Qizhang Li, Yiwen Guo, Xiaochen Yang, Wangmeng Zuo, Hao Chen
Adversarial Example Fine Tuning Multi Step Adversarial Attack Improving Transferability Transfer Based Attack

April 14, 2023

Generating Adversarial Examples with Better Transferability via Masking Unimportant Parameters of Surrogate Model
Dingcheng Yang, Wenjian Yu, Zihao Xiao, Jiaqi Luo
Adversarial Example Surrogate Model Task Transferability Importance Score Transfer Based Attack Transfer Based Adversarial Attack

March 28, 2023

TransAudio: Towards the Transferable Adversarial Audio Attack via Learning Contextualized Perturbations
Qi Gege, Yuefeng Chen, Xiaofeng Mao, Yao Zhu, Binyuan Hui, Xiaodan Li, Rong Zhang, Hui Xue
Adversarial Example Transferable Attack Adversarial Audio Word Level Attack Transfer Based Attack Context Perturbation

March 10, 2023

Boosting Adversarial Attacks by Leveraging Decision Boundary Information
Boheng Zeng, LianLi Gao, QiLong Zhang, ChaoQun Li, JingKuan Song, ShuaiQi Jing
Adversarial Attack Adversarial Example Decision Boundary Transfer Based Attack Boundary Attack

November 23, 2022

Query Efficient Cross-Dataset Transferable Black-Box Attack on Action Recognition
Rohit Gupta, Naveed Akhtar, Gaurav Kumar Nayak, Ajmal Mian, Mubarak Shah
Adversarial Attack Action Recognition Black Box Black Box Attack Efficient Query Transfer Based Attack

Transfer Based Attack

Papers

Downstream Transfer Attack: Adversarial Attacks on Downstream Models with Pre-trained Vision Transformers

Properties that allow or prohibit transferability of adversarial attacks among quantized networks

Towards Evaluating Transfer-based Attacks Systematically, Practically, and Fairly

Structure Invariant Transformation for better Adversarial Transferability

Boosting Adversarial Transferability by Block Shuffle and Rotation

Backpropagation Path Search On Adversarial Transferability

Improving Transferability of Adversarial Examples via Bayesian Attacks

Generating Adversarial Examples with Better Transferability via Masking Unimportant Parameters of Surrogate Model

TransAudio: Towards the Transferable Adversarial Audio Attack via Learning Contextualized Perturbations

Boosting Adversarial Attacks by Leveraging Decision Boundary Information

Query Efficient Cross-Dataset Transferable Black-Box Attack on Action Recognition