Latest Articles for Cybersecurity

A new method for detecting vehicle misbehavior while protecting privacy.

A study on improving video stream detection while maintaining user privacy.

A method to analyze network telescope data for identifying coordinated computer activities.

SPSG improves model stealing efficiency using superpixels and sample gradients.

A new method improves detection of audio deepfakes using similar sample references.

A new approach enhances intrusion detection in decentralized systems.

A look into the risks and defenses in 3D printing.

New research reveals gaps in detecting transient domains used for online abuse.

A new method to improve adversarial training for deep neural networks.

FV8 enhances detection of evasive techniques in JavaScript through dynamic analysis.

A new method combines PUFs and protocols for secure IoT device communication.

This paper discusses securing smart contracts through a type system focused on integrity.

This paper presents EFRAP, a defense against quantization-conditioned backdoor attacks in deep learning models.

A look into the reliability and vulnerabilities of PUFs in IoT devices.

A new approach improves the security of neural networks against adversarial examples.

A study reveals the WordGame attack, exploiting weaknesses in LLM safety measures.

A major vulnerability in XMPU exposes sensitive data after process termination.

Aiming for efficient energy management through smart grids and new technologies.

This study examines the weaknesses of SER models against adversarial attacks across languages.

A framework to enhance firmware update management for smart inverters.

A method to secure neural networks against vulnerabilities in memory.

Introducing a new way to assess privacy risks in machine learning models.

Exploring automated systems for efficient investigation of cyber threats and breaches.

Quantum authentication provides a secure method for data verification using quantum mechanics.

This article discusses the importance and strategies of securing software supply chains.

WPS can aid location services but threatens user privacy significantly.

Improving machine learning robustness against adversarial examples is critical for secure applications.

A study introduces RepNoise to protect large language models from harmful fine-tuning.

Improving distributed learning through adaptive robust clipping techniques.

A study on improving robustness against attacks in language models.

A new framework enhances the efficiency of automated penetration testing using reinforcement learning.

Exploring the role of AI in fixing software vulnerabilities.

Examining the risks facing TEEs due to evolving attack methods.

Examining the impact of nudges on password change behavior after data breaches.

Exploring AI's impact on efficiency, privacy, and security in vital systems.

Research on how harmful agents can corrupt good agents in decentralized RL.

This article examines how attacks affect LLM safety and response generation.

NCS enables effective adversarial example generation with lower computational costs.

A universal audio clip can mute advanced ASR models like Whisper.

A new method improves jailbreak attacks on advanced AI models using character role-play.