Adversarial Example

Adversarial examples are subtly altered inputs designed to fool machine learning models, primarily deep neural networks (DNNs), into making incorrect predictions. Current research focuses on improving model robustness against these attacks, exploring techniques like ensemble methods, multi-objective representation learning, and adversarial training, often applied to architectures such as ResNets and Vision Transformers. Understanding and mitigating the threat of adversarial examples is crucial for ensuring the reliability and security of AI systems across diverse applications, from image classification and natural language processing to malware detection and autonomous driving. The development of robust defenses and effective attack detection methods remains a significant area of ongoing investigation.

Papers

February 5, 2024

February 4, 2024

DeSparsify: Adversarial Attack Against Token Sparsification Mechanisms in Vision Transformers
Oryan Yehezkel, Alon Zolfi, Amit Baras, Yuval Elovici, Asaf Shabtai
Adversarial Attack Vision Transformer Adversarial Example Token Sparsification

February 3, 2024

Seeing is not always believing: The Space of Harmless Perturbations
Lu Chen, Shaofeng Li, Benhao Huang, Fan Yang, Zheng Li, Jie Li, Yuan Luo
Deep Neural Network Adversarial Example Deep Space Plain Sight High OOD Robustness Unintentional Perturbation

February 2, 2024

January 30, 2024

January 24, 2024

Boosting the Transferability of Adversarial Examples via Local Mixup and Adaptive Step Size
Junlin Liu, Xinchen Lyu
Adversarial Example Task Transferability Adversarial Image Step Size Adversarial Framework Adversarial Generation Local Mixup

January 21, 2024

January 19, 2024

January 18, 2024

Adapters Mixup: Mixing Parameter-Efficient Adapters to Enhance the Adversarial Robustness of Fine-tuned Pre-trained Text Classifiers
Tuc Nguyen, Thai Le
Adversarial Attack Adversarial Example Adversarial Training Adversarial Robustness Adapter Fusion Adversarial Augmentation Parameter Efficient Adapter

January 17, 2024

Rethinking Impersonation and Dodging Attacks on Face Recognition Systems
Fengfan Zhou, Qianyu Zhou, Bangjie Yin, Hui Zheng, Xuequan Lu, Lizhuang Ma, Hefei Ling
Adversarial Attack Adversarial Example Adversarial Perturbation Context Impersonation

Adversarial Example

Papers

Arabic Synonym BERT-based Adversarial Examples for Text Classification

Transcending Adversarial Perturbations: Manifold-Aided Adversarial Examples with Legitimate Semantics

A Generative Approach to Surrogate-based Black-box Attacks

DeSparsify: Adversarial Attack Against Token Sparsification Mechanisms in Vision Transformers

Seeing is not always believing: The Space of Harmless Perturbations

$σ$-zero: Gradient-based Optimization of $\ell_0$-norm Adversarial Examples

HQA-Attack: Toward High Quality Black-Box Hard-Label Adversarial Attack on Text

STAA-Net: A Sparse and Transferable Adversarial Attack for Speech Emotion Recognition

Improving QA Model Performance with Cartographic Inoculation

Single Word Change is All You Need: Designing Attacks and Defenses for Text Classifiers

Boosting the Transferability of Adversarial Examples via Local Mixup and Adaptive Step Size

How Robust Are Energy-Based Models Trained With Equilibrium Propagation?

Finding a Needle in the Adversarial Haystack: A Targeted Paraphrasing Approach For Uncovering Edge Cases with Minimal Distribution Distortion

The Surprising Harmfulness of Benign Overfitting for Adversarial Robustness

PuriDefense: Randomized Local Implicit Adversarial Purification for Defending Black-box Query-based Attacks

Adapters Mixup: Mixing Parameter-Efficient Adapters to Enhance the Adversarial Robustness of Fine-tuned Pre-trained Text Classifiers

Rethinking Impersonation and Dodging Attacks on Face Recognition Systems

Robust Localization of Key Fob Using Channel Impulse Response of Ultra Wide Band Sensors for Keyless Entry Systems

Bag of Tricks to Boost Adversarial Transferability

A Generative Adversarial Attack for Multilingual Text Classifiers