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

March 16, 2024

Understanding Robustness of Visual State Space Models for Image Classification
Chengbin Du, Yanxi Li, Chang Xu
Adversarial Attack Native Robustness Adversarial Robustness Image Classification Adversarial Patch Attack Visual State Space Model General Robustness
Improving Adversarial Transferability of Visual-Language Pre-training Models through Collaborative Multimodal Interaction
Jiyuan Fu, Zhaoyu Chen, Kaixun Jiang, Haijing Guo, Jiafeng Wang, Shuyong Gao, Wenqiang Zhang
Adversarial Attack Full Model Adversarial Robustness Multimodal Interaction Image Gradient Multimodal Attack Modality Interaction

March 15, 2024

Benchmarking Zero-Shot Robustness of Multimodal Foundation Models: A Pilot Study
Chenguang Wang, Ruoxi Jia, Xin Liu, Dawn Song
Adversarial Attack Native Robustness Zero Shot Multimodal Foundation Model Pilot Study Robustness Gap Robustness Evaluation Benchmark
Introducing Adaptive Continuous Adversarial Training (ACAT) to Enhance ML Robustness
Mohamed elShehaby, Aditya Kotha, Ashraf Matrawy
Adversarial Attack Adversarial Training Adversarial Sample Machine Learning Robustness
Revisiting Adversarial Training under Long-Tailed Distributions
Xinli Yue, Ningping Mou, Qian Wang, Lingchen Zhao
Adversarial Attack Adversarial Training Long Tailed Distribution Robust Overfitting Balanced Softmax

March 14, 2024

Robust Subgraph Learning by Monitoring Early Training Representations
Sepideh Neshatfar, Salimeh Yasaei Sekeh
Adversarial Attack Adversarial Robustness Graph Learning Pre Trained Representation Input Graph Robust Graph
Towards White Box Deep Learning
Maciej Satkiewicz
Deep Neural Network Adversarial Attack Adversarial Training Semantic Feature Adversarial Testing
What Sketch Explainability Really Means for Downstream Tasks
Hmrishav Bandyopadhyay, Pinaki Nath Chowdhury, Ayan Kumar Bhunia, Aneeshan Sain, Tao Xiang, Yi-Zhe Song
Adversarial Attack High Explainability Downstream Task Stroke Level
Adversarial Fine-tuning of Compressed Neural Networks for Joint Improvement of Robustness and Efficiency
Hallgrimur Thorsteinsson, Valdemar J Henriksen, Tong Chen, Raghavendra Selvan
Neural Network Adversarial Attack Native Robustness High Efficiency Adversarial Training Adversarial Robustness Model Robustness Adversarial Fine Tuning
Adversarial Training with OCR Modality Perturbation for Scene-Text Visual Question Answering
Zhixuan Shen, Haonan Luo, Sijia Li, Tianrui Li
Adversarial Attack Adversarial Training Optical Character Recognition
Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement
Daiwei Yu, Zhuorong Li, Lina Wei, Canghong Jin, Yun Zhang, Sixian Chan
Adversarial Attack Adversarial Training Adversarial Robustness Robust Generalization Label Refinement Robust Overfitting

March 13, 2024

Verifix: Post-Training Correction to Improve Label Noise Robustness with Verified Samples
Sangamesh Kodge, Deepak Ravikumar, Gobinda Saha, Kaushik Roy
Adversarial Attack Label Noise Language Correction Training Dynamic Label Corruption Label Noise Robustness
Versatile Defense Against Adversarial Attacks on Image Recognition
Haibo Zhang, Zhihua Yao, Kouichi Sakurai
Adversarial Attack Image Recognition Attack Paradigm

March 11, 2024

epsilon-Mesh Attack: A Surface-based Adversarial Point Cloud Attack for Facial Expression Recognition
Batuhan Cengiz, Mert Gulsen, Yusuf H. Sahin, Gozde Unal
Adversarial Attack Point Cloud Facial Expression Recognition Mesh Attack

March 9, 2024

Hard-label based Small Query Black-box Adversarial Attack
Jeonghwan Park, Paul Miller, Niall McLaughlin
Adversarial Attack Black Box Attack Method Hard Label Limited Query
IOI: Invisible One-Iteration Adversarial Attack on No-Reference Image- and Video-Quality Metrics
Ekaterina Shumitskaya, Anastasia Antsiferova, Dmitriy Vatolin
Adversarial Attack Reference Image Video Benchmark Video Quality Metric Replay Attack Iteration Head

March 8, 2024

The Impact of Quantization on the Robustness of Transformer-based Text Classifiers
Seyed Parsa Neshaei, Yasaman Boreshban, Gholamreza Ghassem-Sani, Seyed Abolghasem Mirroshandel
Adversarial Attack Native Robustness Transformer Based Adversarial Training Global Impact Transformer Based Model Quantization Operator Adversarial Evaluation
Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds
Tianrui Lou, Xiaojun Jia, Jindong Gu, Li Liu, Siyuan Liang, Bangyan He, Xiaochun Cao
Adversarial Attack Adversarial Perturbation 3D Point Cloud Imperceptible Pattern Physical Adversarial Adversarial Shape Prior
Exploring the Adversarial Frontier: Quantifying Robustness via Adversarial Hypervolume
Ping Guo, Cheng Gong, Xi Lin, Zhiyuan Yang, Qingfu Zhang
Adversarial Attack Native Robustness Adversarial Robustness Adversarial Learning Adversarial Threat
Defending Against Unforeseen Failure Modes with Latent Adversarial Training
Stephen Casper, Lennart Schulze, Oam Patel, Dylan Hadfield-Menell
Adversarial Attack Adversarial Training Failure Mode Novel Attack Attack Surface

Adversarial Attack

Papers

Understanding Robustness of Visual State Space Models for Image Classification

Improving Adversarial Transferability of Visual-Language Pre-training Models through Collaborative Multimodal Interaction

Benchmarking Zero-Shot Robustness of Multimodal Foundation Models: A Pilot Study

Introducing Adaptive Continuous Adversarial Training (ACAT) to Enhance ML Robustness

Revisiting Adversarial Training under Long-Tailed Distributions

Robust Subgraph Learning by Monitoring Early Training Representations

Towards White Box Deep Learning

What Sketch Explainability Really Means for Downstream Tasks

Adversarial Fine-tuning of Compressed Neural Networks for Joint Improvement of Robustness and Efficiency

Adversarial Training with OCR Modality Perturbation for Scene-Text Visual Question Answering

Soften to Defend: Towards Adversarial Robustness via Self-Guided Label Refinement

Verifix: Post-Training Correction to Improve Label Noise Robustness with Verified Samples

Versatile Defense Against Adversarial Attacks on Image Recognition

epsilon-Mesh Attack: A Surface-based Adversarial Point Cloud Attack for Facial Expression Recognition

Hard-label based Small Query Black-box Adversarial Attack

IOI: Invisible One-Iteration Adversarial Attack on No-Reference Image- and Video-Quality Metrics

The Impact of Quantization on the Robustness of Transformer-based Text Classifiers

Hide in Thicket: Generating Imperceptible and Rational Adversarial Perturbations on 3D Point Clouds

Exploring the Adversarial Frontier: Quantifying Robustness via Adversarial Hypervolume

Defending Against Unforeseen Failure Modes with Latent Adversarial Training