Latest Articles for Adversarial Attacks

A framework to evaluate the robustness of Bayesian Neural Networks against adversarial attacks.

This study analyzes image distribution using generative models for better recognition.

NatLogAttack improves adversarial testing of language models using natural logic.

Exploring the challenges and defenses in quantum adversarial machine learning.

Examining how single-class attacks can mislead deep learning models.

New method improves adversarial attacks on deep neural networks using semantic changes.

This article examines the security risks of Non-intrusive Load Monitoring systems.

New methods enhance training against adversarial attacks by focusing on example vulnerabilities.

Examining the strengths and weaknesses of split learning in data privacy.

Examining the risks and explainability challenges of adversarial attacks on AI models.

A fresh approach to tackling vulnerabilities in mixed classifiers.

Deep learning faces serious risks from adversarial attacks that mislead systems.

Examining the characteristics and security risks of on-device models in iOS apps.

New method enhances adversarial examples against language model vulnerabilities.

Investigating vulnerabilities of DNNs against adversarial noise in image denoising.

A study on SCS versus traditional convolutional layers in image classification.

New method improves adversarial attacks while maintaining performance metrics.

Explore strategies to enhance information sharing in large networks.

A new method improves image classification by combining multiple self-supervised tasks.

Examining challenges and defenses against adversarial attacks on 3D point clouds.

Research shows how mimicking human vision can strengthen deep neural networks against attacks.

New methods improve object detection systems' resistance to adversarial changes.

Examining the threats posed by adversarial attacks on AI systems.

New methods to secure federated learning against attacks while preserving user privacy.

Focusing on training data selection can strengthen node classification systems.

New techniques improve the reliability of medical imaging tools against harmful data alterations.

Explore the impact of adversarial attacks on multi-agent reinforcement learning systems.

DyNNs provide efficient processing and robust performance against adversarial attacks.

Examining how model compression affects defenses against adversarial attacks.

A new framework improves deep learning model reliability against subtle changes.

Examining how adversarial attacks impact DeepReceiver in wireless communication systems.

Research highlights new ways to improve model defenses against adversarial attacks.

Researchers reveal how malware can deceive antivirus through clever attacks.

A new method enhances the robustness of Mixture of Experts CNNs against adversarial inputs.

A novel framework improves the effectiveness of adversarial attacks on code models.

Examining the dangers posed by adversarial attacks on image and text models.

Examining how Causal Neural Networks enhance resilience to adversarial attacks.

This article explores the role of Transformers in image restoration and their vulnerabilities to adversarial attacks.

Using perplexity to identify risky inputs in language models.

Examining the risks associated with leading computer vision models and their effectiveness.