Adversarial Robustness

Adversarial robustness focuses on developing machine learning models resistant to adversarial attacks—small, carefully crafted input perturbations designed to cause misclassification. Current research investigates diverse defense mechanisms, including adversarial training, data purification using diffusion models, and biologically-inspired regularizers, often applied to convolutional neural networks (CNNs), transformers, and spiking neural networks (SNNs). This field is crucial for ensuring the reliability and safety of AI systems in real-world applications, particularly in safety-critical domains like autonomous driving and healthcare, where model failures can have severe consequences.

Papers

November 18, 2024

Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods
Egor Kovalev, Georgii Bychkov, Khaled Abud, Aleksandr Gushchin, Anna Chistyakova, Sergey Lavrushkin, Dmitriy Vatolin, Anastasia Antsiferova
Adversarial Attack Adversarial Robustness NCD Method Consistent Comparison New Method Neural Image Compression

November 15, 2024

Edge-Only Universal Adversarial Attacks in Distributed Learning
Giulio Rossolini, Tommaso Baldi, Alessandro Biondi, Giorgio Buttazzo
Adversarial Robustness Distributed Learning Universal Adversarial Adversarial Influence

November 14, 2024

BEARD: Benchmarking the Adversarial Robustness for Dataset Distillation
Zheng Zhou, Wenquan Feng, Shuchang Lyu, Guangliang Cheng, Xiaowei Huang, Qi Zhao
Adversarial Attack Adversarial Robustness Dataset Distillation Adversarial Game Robustness Measurement Facial Hair

November 13, 2024

Confidence-aware Denoised Fine-tuning of Off-the-shelf Models for Certified Robustness
Suhyeok Jang, Seojin Kim, Jinwoo Shin, Jongheon Jeong
Native Robustness Adversarial Robustness Certified Robustness Confidence Aware Image Denoiser Shelf Model

November 12, 2024

Rapid Response: Mitigating LLM Jailbreaks with a Few Examples
Alwin Peng, Julian Michael, Henry Sleight, Ethan Perez, Mrinank Sharma
Adversarial Robustness Jailbreak Attack Robust Defense

November 9, 2024

AI-Compass: A Comprehensive and Effective Multi-module Testing Tool for AI Systems
Zhiyu Zhu, Zhibo Jin, Hongsheng Hu, Minhui Xue, Ruoxi Sun, Seyit Camtepe, Praveen Gauravaram, Huaming Chen
Adversarial Attack Adversarial Robustness AI System

November 7, 2024

November 6, 2024

A Fundamental Accuracy--Robustness Trade-off in Regression and Classification
Sohail Bahmani
Classification Code Adversarial Robustness Novel Regression Adversarial Risk Accuracy Robustness Ridge Function

November 5, 2024

Enhancing Adversarial Robustness via Uncertainty-Aware Distributional Adversarial Training
Junhao Dong, Xinghua Qu, Z. Jane Wang, Yew-Soon Ong
Adversarial Example Adversarial Training Adversarial Robustness Domain Adversarial

October 30, 2024

FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Training
Tejaswini Medi, Steffen Jung, Margret Keuper
Adversarial Attack Adversarial Training Adversarial Robustness Model Fairness Fair Adversarial

October 29, 2024

October 26, 2024

Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions
Poojitha Thota, Shirin Nilizadeh
Adversarial Robustness Adversarial Perturbation New Attack Abstractive Summary Influence Function Abstractive Summarization Model Systematic Bias Text Summary

October 24, 2024

Complexity Matters: Effective Dimensionality as a Measure for Adversarial Robustness
David Khachaturov, Robert Mullins
Native Robustness Adversarial Robustness Complexity Matter Robustness Evaluation Effective Dimension Robustness Measurement

October 21, 2024

Conflict-Aware Adversarial Training
Zhiyu Xue, Haohan Wang, Yao Qin, Ramtin Pedarsani
Adversarial Training Adversarial Robustness Adversarial Sample Adversarial Loss

October 19, 2024

October 18, 2024

A Hybrid Defense Strategy for Boosting Adversarial Robustness in Vision-Language Models
Yuhan Liang, Yijun Li, Yumeng Niu, Qianhe Shen, Hangyu Liu
Adversarial Attack Vision Language Model Adversarial Example Adversarial Training Adversarial Robustness CLIP Model

October 16, 2024

New Paradigm of Adversarial Training: Breaking Inherent Trade-Off between Accuracy and Robustness via Dummy Classes
Yanyun Wang, Li Liu, Zi Liang, Qingqing Ye, Haibo Hu
Native Robustness Adversarial Example Adversarial Training Adversarial Robustness Adversarial Sample Robust Learning Function Class

Adversarial Robustness

Papers

Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods

Edge-Only Universal Adversarial Attacks in Distributed Learning

BEARD: Benchmarking the Adversarial Robustness for Dataset Distillation

Confidence-aware Denoised Fine-tuning of Off-the-shelf Models for Certified Robustness

Rapid Response: Mitigating LLM Jailbreaks with a Few Examples

AI-Compass: A Comprehensive and Effective Multi-module Testing Tool for AI Systems

Adversarial Robustness of In-Context Learning in Transformers for Linear Regression

Game-Theoretic Defenses for Robust Conformal Prediction Against Adversarial Attacks in Medical Imaging

A Fundamental Accuracy--Robustness Trade-off in Regression and Classification

Enhancing Adversarial Robustness via Uncertainty-Aware Distributional Adversarial Training

FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Training

CausAdv: A Causal-based Framework for Detecting Adversarial Examples

Text-Guided Attention is All You Need for Zero-Shot Robustness in Vision-Language Models

Attacks against Abstractive Text Summarization Models through Lead Bias and Influence Functions

Complexity Matters: Effective Dimensionality as a Measure for Adversarial Robustness

Conflict-Aware Adversarial Training

Beyond Pruning Criteria: The Dominant Role of Fine-Tuning and Adaptive Ratios in Neural Network Robustness

Toward Robust RALMs: Revealing the Impact of Imperfect Retrieval on Retrieval-Augmented Language Models

A Hybrid Defense Strategy for Boosting Adversarial Robustness in Vision-Language Models

New Paradigm of Adversarial Training: Breaking Inherent Trade-Off between Accuracy and Robustness via Dummy Classes