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

May 22, 2023

FGAM:Fast Adversarial Malware Generation Method Based on Gradient Sign
Kun Li, Fan Zhang, Wei Guo
Adversarial Attack Adversarial Malware

May 21, 2023

Are Your Explanations Reliable? Investigating the Stability of LIME in Explaining Text Classifiers by Marrying XAI and Adversarial Attack
Christopher Burger, Lingwei Chen, Thai Le
Adversarial Attack Explainable AI Line by Line Explanation Core Stability XAI Method Text Classifier Text Perturbation Dseg Lime Text Semantics Explanation Selection

May 19, 2023

DAP: A Dynamic Adversarial Patch for Evading Person Detectors
Amira Guesmi, Ruitian Ding, Muhammad Abdullah Hanif, Ihsen Alouani, Muhammad Shafique
Adversarial Attack Generative Adversarial Network Adversarial Patch Image Patch Person Detector Physical Attack

May 18, 2023

May 17, 2023

May 16, 2023

Ortho-ODE: Enhancing Robustness and of Neural ODEs against Adversarial Attacks
Vishal Purohit
Adversarial Attack Native Robustness Deep Learning Model Neural ODE Neural Ordinary Differential Equation Intrinsic Robustness High Order Solver

May 15, 2023

Attacking Perceptual Similarity Metrics
Abhijay Ghildyal, Feng Liu
Adversarial Attack Adversarial Perturbation Imperceptible Adversarial Perturbation Perceptual Similarity Perceptual Similarity Metric

Adversarial Attack

Papers

FGAM:Fast Adversarial Malware Generation Method Based on Gradient Sign

Are Your Explanations Reliable? Investigating the Stability of LIME in Explaining Text Classifiers by Marrying XAI and Adversarial Attack

Dynamic Transformers Provide a False Sense of Efficiency

SneakyPrompt: Jailbreaking Text-to-image Generative Models

Multi-Task Models Adversarial Attacks

DAP: A Dynamic Adversarial Patch for Evading Person Detectors

Quantifying the robustness of deep multispectral segmentation models against natural perturbations and data poisoning

Attacks on Online Learners: a Teacher-Student Analysis

Deep PackGen: A Deep Reinforcement Learning Framework for Adversarial Network Packet Generation

How Deep Learning Sees the World: A Survey on Adversarial Attacks & Defenses

Towards an Accurate and Secure Detector against Adversarial Perturbations

Adversarial Amendment is the Only Force Capable of Transforming an Enemy into a Friend

Content-based Unrestricted Adversarial Attack

Exact Recovery for System Identification with More Corrupt Data than Clean Data

Raising the Bar for Certified Adversarial Robustness with Diffusion Models

Ortho-ODE: Enhancing Robustness and of Neural ODEs against Adversarial Attacks

Attacking Perceptual Similarity Metrics

Diffusion Models for Imperceptible and Transferable Adversarial Attack

Improving Defensive Distillation using Teacher Assistant

On enhancing the robustness of Vision Transformers: Defensive Diffusion