Latest Articles for Cybersecurity

This article discusses the risks of package hallucinations in code-generating models.

Using language models to improve software bug detection and testing efficiency.

New benchmark CTIBench evaluates LLM effectiveness in tackling cyber threats.

A look at kernel safety methods to combat advanced attacks like Spectre.

A new digital signature system improves speed and security in data center applications.

Examining the Rowhammer flaw and its implications for data security.

Investigating the security flaws in Memory Tagging Extension and speculative execution attacks.

Examining the security risks and solutions for emerging AI assistant technologies.

A look at vulnerabilities and solutions for deep learning systems.

FSBI method improves detection of manipulated digital media.

A new tool improves static analysis with simplified graphs and machine learning.

Competition reveals vulnerabilities and defenses in language model security.

A method to identify attacks on systems combining images and text.

New method targets rhythm changes for stealthy speech attacks.

New methods improve unlearning harmful data in machine learning systems.

A look into robust learning models and their importance in data security.

Examining security risks in mmWave networks utilizing simultaneous wireless information and power transfer.

A study on improving model extraction techniques for deep learning security.

A new approach combines knowledge and technology to improve software vulnerability detection.

This paper introduces a framework to evaluate vulnerabilities in large language models.

Centaurs combine human insight and AI to improve decision-making across various fields.

A new method for encrypted control enhances security while performing necessary functions.

A plan to bolster the electricity grid's resilience and reliability amid emerging digital risks.

This article explores the impact of data poisoning on language model alignment.

P3GNN enhances APT detection while protecting data privacy in SDN networks.

Learn how backdoor attacks threaten machine learning systems and methods to defend against them.

Tracemax offers improved packet tracing for better DDoS attack defense.

A framework to improve APT detection while protecting privacy.

Exploring the safety challenges posed by adversarial attacks on multimodal agents.

Addressing the safety concerns of code produced by large language models.

LLMs can aid in social engineering protection and also pose new risks.

A new defense strategy for LLMs against backdoor attacks.

Cybersecurity evolves with zero-trust networking to combat rising threats.

A new lightweight approach to enhance cyber threat sharing among IoT devices.

DF40 offers a comprehensive approach to improving deepfake detection methods.

SeCTIS enables safe sharing of cyber threat intelligence while maintaining privacy and data quality.

This article discusses the security risks and defense strategies for large language models.

A study on how language models can ignore instructions and their implications.

QuADTool simplifies risk assessment through attack-defense trees.

New techniques enhance privacy and security in database queries and transactions.