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

October 30, 2024

October 28, 2024

October 27, 2024

October 26, 2024

October 24, 2024

Adversarial Attacks on Large Language Models Using Regularized Relaxation
Samuel Jacob Chacko, Sajib Biswas, Chashi Mahiul Islam, Fatema Tabassum Liza, Xiuwen Liu
Large Language Model Language Model Adversarial Attack Adversarial Input Confidence Relaxation

October 23, 2024

Advancing NLP Security by Leveraging LLMs as Adversarial Engines
Sudarshan Srinivasan, Maria Mahbub, Amir Sadovnik
Large Language Model Adversarial Attack Adversarial Learning Adversarial Natural Language

Adversarial Attack

Papers

Protecting Feed-Forward Networks from Adversarial Attacks Using Predictive Coding

Adversarial Attacks of Vision Tasks in the Past 10 Years: A Survey

Rethinking Inverse Reinforcement Learning: from Data Alignment to Task Alignment

Keep on Swimming: Real Attackers Only Need Partial Knowledge of a Multi-Model System

FAIR-TAT: Improving Model Fairness Using Targeted Adversarial Training

Effective and Efficient Adversarial Detection for Vision-Language Models via A Single Vector

Automated Trustworthiness Oracle Generation for Machine Learning Text Classifiers

Longitudinal Mammogram Exam-based Breast Cancer Diagnosis Models: Vulnerability to Adversarial Attacks

Embedding-based classifiers can detect prompt injection attacks

Enhancing Adversarial Attacks through Chain of Thought

AdvI2I: Adversarial Image Attack on Image-to-Image Diffusion models

SeriesGAN: Time Series Generation via Adversarial and Autoregressive Learning

Resilience in Knowledge Graph Embeddings

Evaluating the Robustness of LiDAR Point Cloud Tracking Against Adversarial Attack

LLM Robustness Against Misinformation in Biomedical Question Answering

Integrating uncertainty quantification into randomized smoothing based robustness guarantees

Transferable Adversarial Attacks on SAM and Its Downstream Models

Generative Adversarial Patches for Physical Attacks on Cross-Modal Pedestrian Re-Identification

Adversarial Attacks on Large Language Models Using Regularized Relaxation

Advancing NLP Security by Leveraging LLMs as Adversarial Engines