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 26, 2023

DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution
Matías P. Pizarro B., Dorothea Kolossa, Asja Fischer
Adversarial Attack Automatic Speech Recognition Adversarial Example Automatic Speech Recognition System Distinct AInality Trait Adversarial Data Adversarial Audio Output Distribution
Trust-Aware Resilient Control and Coordination of Connected and Automated Vehicles
H M Sabbir Ahmad, Ehsan Sabouni, Wei Xiao, Christos G. Cassandras, Wenchao Li
Adversarial Attack Automated Driving Traffic Network Prior Coordination Connected Behavior Resilient Control

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
Adversarial Attacks on Leakage Detectors in Water Distribution Networks
Paul Stahlhofen, André Artelt, Luca Hermes, Barbara Hammer
Adversarial Attack Model Robustness Water Distribution Network Leakage Detector
IDEA: Invariant Defense for Graph Adversarial Robustness
Shuchang Tao, Qi Cao, Huawei Shen, Yunfan Wu, Bingbing Xu, Xueqi Cheng
Adversarial Attack Adversarial Example Adversarial Robustness Adversarial Graph
PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentation
Xianghao Jiao, Yaohua Liu, Jiaxin Gao, Xinyuan Chu, Risheng Liu, Xin Fan
Adversarial Attack Semantic Segmentation Adversarial Perturbation Image Deraining Adversarial Prediction Robust Adversarial Learning

May 24, 2023

May 23, 2023

The Best Defense is a Good Offense: Adversarial Augmentation against Adversarial Attacks
Iuri Frosio, Jan Kautz
Adversarial Attack Adversarial Augmentation
QFA2SR: Query-Free Adversarial Transfer Attacks to Speaker Recognition Systems
Guangke Chen, Yedi Zhang, Zhe Zhao, Fu Song
Adversarial Attack Black Box Attack Voice Assistant Speaker Recognition System Transfer Based Adversarial Attack
Expressive Losses for Verified Robustness via Convex Combinations
Alessandro De Palma, Rudy Bunel, Krishnamurthy Dvijotham, M. Pawan Kumar, Robert Stanforth, Alessio Lomuscio
Adversarial Attack Adversarial Training Adversarial Robustness Expressive Power Robustness Verification Accuracy Robustness Convex Combination
Enhancing Accuracy and Robustness through Adversarial Training in Class Incremental Continual Learning
Minchan Kwon, Kangil Kim
Adversarial Attack Native Robustness Adversarial Example Adversarial Training Class Incremental Continual Learning Enhancing Accuracy
DiffProtect: Generate Adversarial Examples with Diffusion Models for Facial Privacy Protection
Jiang Liu, Chun Pong Lau, Rama Chellappa
Diffusion Model Adversarial Attack Adversarial Example Face Image Privacy Protection Facial Recognition System

May 22, 2023

Adversarial Nibbler: A Data-Centric Challenge for Improving the Safety of Text-to-Image Models
Alicia Parrish, Hannah Rose Kirk, Jessica Quaye, Charvi Rastogi, Max Bartolo, Oana Inel, Juan Ciro, Rafael Mosquera, Addison Howard, Will Cukierski, D. Sculley, Vijay Janapa Reddi, Lora Aroyo
Adversarial Attack Text to Image Model Human SAFETY Generative AI Model Adversarial Challenge
Latent Magic: An Investigation into Adversarial Examples Crafted in the Semantic Latent Space
BoYang Zheng
Adversarial Attack Adversarial Example Latent Space Comprehensive Investigation LangId Magic Spell
Towards Benchmarking and Assessing Visual Naturalness of Physical World Adversarial Attacks
Simin Li, Shuing Zhang, Gujun Chen, Dong Wang, Pu Feng, Jiakai Wang, Aishan Liu, Xin Yi, Xianglong Liu
Adversarial Attack Benchmark Platform Visual Naturalness Physical World Attack Naturalness Assessment Benchmark Attack Enhancement Attack
Flying Adversarial Patches: Manipulating the Behavior of Deep Learning-based Autonomous Multirotors
Pia Hanfeld, Marina M. -C. Höhne, Michael Bussmann, Wolfgang Hönig
Adversarial Attack BEHAVIOR Explanation Adversarial Patch Autonomous Quadrotor
Uncertainty-based Detection of Adversarial Attacks in Semantic Segmentation
Kira Maag, Asja Fischer
Adversarial Attack Semantic Segmentation Adversarial Example Segmentation Result Semantic Image Segmentation Detection of Uncertainty in Exceedance
The defender's perspective on automatic speaker verification: An overview
Haibin Wu, Jiawen Kang, Lingwei Meng, Helen Meng, Hung-yi Lee
Adversarial Attack Speaker Verification Visual Perspective Synthesized Speech Fake Speech RePLAy Loss Multiple Defender

Adversarial Attack

Papers

DistriBlock: Identifying adversarial audio samples by leveraging characteristics of the output distribution

Trust-Aware Resilient Control and Coordination of Connected and Automated Vehicles

Don't Retrain, Just Rewrite: Countering Adversarial Perturbations by Rewriting Text

Adversarial Attacks on Leakage Detectors in Water Distribution Networks

IDEA: Invariant Defense for Graph Adversarial Robustness

PEARL: Preprocessing Enhanced Adversarial Robust Learning of Image Deraining for Semantic Segmentation

How do humans perceive adversarial text? A reality check on the validity and naturalness of word-based adversarial attacks

Relating Implicit Bias and Adversarial Attacks through Intrinsic Dimension

AdvFunMatch: When Consistent Teaching Meets Adversarial Robustness

The Best Defense is a Good Offense: Adversarial Augmentation against Adversarial Attacks

QFA2SR: Query-Free Adversarial Transfer Attacks to Speaker Recognition Systems

Expressive Losses for Verified Robustness via Convex Combinations

Enhancing Accuracy and Robustness through Adversarial Training in Class Incremental Continual Learning

DiffProtect: Generate Adversarial Examples with Diffusion Models for Facial Privacy Protection

Adversarial Nibbler: A Data-Centric Challenge for Improving the Safety of Text-to-Image Models

Latent Magic: An Investigation into Adversarial Examples Crafted in the Semantic Latent Space

Towards Benchmarking and Assessing Visual Naturalness of Physical World Adversarial Attacks

Flying Adversarial Patches: Manipulating the Behavior of Deep Learning-based Autonomous Multirotors

Uncertainty-based Detection of Adversarial Attacks in Semantic Segmentation

The defender's perspective on automatic speaker verification: An overview