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

January 15, 2025

Improving Stability Estimates in Adversarial Explainable AI through Alternate Search Methods
Christopher Burger, Charles Walter
Adversarial Attack High Explainability Learning Problem Complex Model Local Surrogate

January 13, 2025

January 12, 2025

Measuring the Robustness of Reference-Free Dialogue Evaluation Systems
Justin Vasselli, Adam Nohejl, Taro Watanabe
Adversarial Attack Native Robustness Dialogue System Dialogue Evaluation Adversarial Response

January 10, 2025

January 9, 2025

January 8, 2025

January 7, 2025

January 5, 2025

Tougher Text, Smarter Models: Raising the Bar for Adversarial Defence Benchmarks
Yang Wang, Chenghua Lin
Adversarial Attack Deep Learning Model New Benchmark Adversarial Robustness Natural Language Inference AI Model Textual Adversarial Attack Four Bar

January 4, 2025

Adversarial Attack

Papers

Improving Stability Estimates in Adversarial Explainable AI through Alternate Search Methods

Cross-Modal Transferable Image-to-Video Attack on Video Quality Metrics

Towards an End-to-End (E2E) Adversarial Learning and Application in the Physical World

VENOM: Text-driven Unrestricted Adversarial Example Generation with Diffusion Models

MOS-Attack: A Scalable Multi-objective Adversarial Attack Framework

Protego: Detecting Adversarial Examples for Vision Transformers via Intrinsic Capabilities

Measuring the Robustness of Reference-Free Dialogue Evaluation Systems

Effective faking of verbal deception detection with target-aligned adversarial attacks

UV-Attack: Physical-World Adversarial Attacks for Person Detection via Dynamic-NeRF-based UV Mapping

Enforcing Fundamental Relations via Adversarial Attacks on Input Parameter Correlations

CROPS: Model-Agnostic Training-Free Framework for Safe Image Synthesis with Latent Diffusion Models

DiffAttack: Diffusion-based Timbre-reserved Adversarial Attack in Speaker Identification

On Measuring Unnoticeability of Graph Adversarial Attacks: Observations, New Measure, and Applications

LayerMix: Enhanced Data Augmentation through Fractal Integration for Robust Deep Learning

Towards Fair Class-wise Robustness: Class Optimal Distribution Adversarial Training

Synthetic Data Privacy Metrics

Rethinking Adversarial Attacks in Reinforcement Learning from Policy Distribution Perspective

Tougher Text, Smarter Models: Raising the Bar for Adversarial Defence Benchmarks

Distillation-Enhanced Physical Adversarial Attacks

Exploring Secure Machine Learning Through Payload Injection and FGSM Attacks on ResNet-50