Better Adversarial Robustness

Improving the adversarial robustness of machine learning models is a crucial research area focused on making these models less susceptible to malicious attacks that manipulate inputs to cause misclassification. Current efforts explore diverse approaches, including novel model architectures like diffusion classifiers and part-based models, refined adversarial training techniques (e.g., trajectory reweighting and curvature regularization), and innovative defense strategies such as dataset distillation and combined transduction/rejection methods. These advancements are vital for enhancing the reliability and security of machine learning systems across various applications, from image recognition to natural language processing.

Papers

August 16, 2024

Efficient Image-to-Image Diffusion Classifier for Adversarial Robustness
Hefei Mei, Minjing Dong, Chang Xu
Adversarial Robustness Image to Image Diffusion Diffusion Framework Better Adversarial Robustness Image Translation Model

July 15, 2024

PartImageNet++ Dataset: Scaling up Part-based Models for Robust Recognition
Xiao Li, Yining Liu, Na Dong, Sitian Qin, Xiaolin Hu
Adversarial Robustness Part Aware Part Based Robust Recognition Better Adversarial Robustness Part Segmentation Annotation

March 15, 2024

Towards Adversarially Robust Dataset Distillation by Curvature Regularization
Eric Xue, Yijiang Li, Haoyang Liu, Yifan Shen, Haohan Wang
Adversarial Robustness Dataset Distillation Adversarial Text Distilled Dataset Standard Adversarial Training Better Adversarial Robustness Curvature Regularization

June 25, 2023

Enhancing Adversarial Training via Reweighting Optimization Trajectory
Tianjin Huang, Shiwei Liu, Tianlong Chen, Meng Fang, Li Shen, Vlaod Menkovski, Lu Yin, Yulong Pei, Mykola Pechenizkiy
Adversarial Training Better Adversarial Robustness Adversarial Weight Perturbation Optimization Trajectory

May 27, 2023

Two Heads are Better than One: Towards Better Adversarial Robustness by Combining Transduction and Rejection
Nils Palumbo, Yang Guo, Xi Wu, Jiefeng Chen, Yingyu Liang, Somesh Jha
Adversarial Attack Adversarial Perturbation Human Head Risky Prompt Rejection Transmission Data Transductive Approach Better Adversarial Robustness

May 25, 2023

Don't Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text
Ashim Gupta, Carter Wood Blum, Temma Choji, Yingjie Fei, Shalin Shah, Alakananda Vempala, Vivek Srikumar
Adversarial Attack Adversarial Robustness Text Modality Adversarial Perturbation Adversarial Input Better Adversarial Robustness

December 5, 2022

Multiple Perturbation Attack: Attack Pixelwise Under Different $\ell_p$-norms For Better Adversarial Performance
Ngoc N. Tran, Anh Tuan Bui, Dinh Phung, Trung Le
Adversarial Attack Adversarial Example \Ell_p$ Norm Perturbation Attack Better Adversarial Robustness ImageNet Attack

November 8, 2022

Preserving Semantics in Textual Adversarial Attacks
David Herel, Hugo Cisneros, Tomas Mikolov
Adversarial Attack Adversarial Example Textual Adversarial Attack Better Adversarial Robustness Semantic Preservation

August 17, 2022

On the Privacy Effect of Data Enhancement via the Lens of Memorization
Xiao Li, Qiongxiu Li, Zhanhao Hu, Xiaolin Hu
Adversarial Robustness Membership Inference Attack Camera Lens PRIvacy Leakage Limited Memorization Privacy Attack Better Adversarial Robustness Data Enhancement

December 7, 2021

Decision-based Black-box Attack Against Vision Transformers via Patch-wise Adversarial Removal
Yucheng Shi, Yahong Han, Yu-an Tan, Xiaohui Kuang
Vision Transformer Adversarial Robustness Decision Based Better Adversarial Robustness Adversarial Removal