Adversarial Attack

Adversarial attacks aim to deceive machine learning models by subtly altering input data, causing misclassifications or other erroneous outputs. Current research focuses on developing more robust models and detection methods, exploring various attack strategies across different model architectures (including vision transformers, recurrent neural networks, and graph neural networks) and data types (images, text, signals, and tabular data). Understanding and mitigating these attacks is crucial for ensuring the reliability and security of AI systems in diverse applications, from autonomous vehicles to medical diagnosis and cybersecurity.

Papers

November 19, 2023

November 18, 2023

Improving Adversarial Transferability by Stable Diffusion
Jiayang Liu, Siyu Zhu, Siyuan Liang, Jie Zhang, Han Fang, Weiming Zhang, Ee-Chien Chang
Adversarial Attack Adversarial Example Stable Diffusion Adversarial Transferability Diffusion Attack

November 15, 2023

November 14, 2023

November 13, 2023

November 12, 2023

Resilient Graph Neural Networks: A Coupled Dynamical Systems Approach
Moshe Eliasof, Davide Murari, Ferdia Sherry, Carola-Bibiane Schönlieb
Graph Neural Network Adversarial Attack Adversarial Perturbation

November 10, 2023

Adversarial Attack

Papers

Adversarial Prompt Tuning for Vision-Language Models

Robust Network Slicing: Multi-Agent Policies, Adversarial Attacks, and Defensive Strategies

Attention-Based Real-Time Defenses for Physical Adversarial Attacks in Vision Applications

Improving Adversarial Transferability by Stable Diffusion

Hijacking Large Language Models via Adversarial In-Context Learning

Breaking Boundaries: Balancing Performance and Robustness in Deep Wireless Traffic Forecasting

Bergeron: Combating Adversarial Attacks through a Conscience-Based Alignment Framework

A Unified Approach to Learning Ising Models: Beyond Independence and Bounded Width

Adversarial Attacks to Reward Machine-based Reinforcement Learning

$DA^3$: A Distribution-Aware Adversarial Attack against Language Models

Alignment is not sufficient to prevent large language models from generating harmful information: A psychoanalytic perspective

The Perception-Robustness Tradeoff in Deterministic Image Restoration

On The Relationship Between Universal Adversarial Attacks And Sparse Representations

Towards Improving Robustness Against Common Corruptions in Object Detectors Using Adversarial Contrastive Learning

An Extensive Study on Adversarial Attack against Pre-trained Models of Code

Hardest Monotone Functions for Evolutionary Algorithms

Multi-agent Attacks for Black-box Social Recommendations

Resilient Graph Neural Networks: A Coupled Dynamical Systems Approach

Transferability Bound Theory: Exploring Relationship between Adversarial Transferability and Flatness

Fight Fire with Fire: Combating Adversarial Patch Attacks using Pattern-randomized Defensive Patches