Latest Articles for Data Security

LightPHE makes it easier to protect sensitive data using homomorphic encryption.

New method enhances synthetic data generation while ensuring privacy protection.

Exploring how blockchain can improve the safety of large language models.

FedTime combines federated learning and local data for improved forecasting while ensuring data privacy.

BackdoorBench offers a unified approach to assess backdoor learning methods in deep neural networks.

A look at the weaknesses in LLMs and strategies for improvement.

A new method for protecting medical data using blockchain and self-sovereign identity.

Examining privacy concerns with task-oriented communication in the upcoming 6G technology.

Innovative techniques protect sensitive speech data while maintaining processing accuracy.

A new method enhances privacy for facial images while retaining quality.

Research improves methods for analyzing lightweight ciphers like Simon and Simeck.

Examining the risks and strategies of model hijacking in federated learning systems.

New methods improve the analysis of code-based cryptography, enhancing security.

A look into prompt leakage risks in large language models and prevention methods.

Examining how AI assistants can respect user privacy while handling tasks.

Larger language models show increased vulnerability to harmful data behaviors.

Discover the potential of Kramers-Kronig detection in secure quantum communication.

Learn how QR codes work and the role of error correction.

New methods improve deep learning models' resistance to adversarial changes.

New methods reduce identity leakage in face recognition while ensuring data privacy.

RAG systems improve operations but introduce serious security vulnerabilities.

Integrating quantum computing and randomized smoothing improves machine learning model security against attacks.

Investigating the effects of PMD on secure data transfer methods.

New methods enhance data privacy while preserving analysis accuracy.

CRISP safeguards cloud applications against rollback attacks, ensuring data integrity and security.

New methods expose vulnerabilities in medical models through backdoor attacks.

A guide to better comprehend differential privacy and its implications for personal data.

A method to verify identities while protecting personal information.

A flexible system improves access to sensitive information for trusted users.

A look at how on-device learning balances app performance with user privacy.

MIA-Tuner aims to address privacy issues in LLM training data.

Collision-Free Balanced Sequences enhance communication systems by preventing interference.

Explore methods and challenges in maintaining secure communications today.

This study highlights the security challenges posed by DRAM data persistence in FPGA systems.

Validation and confirmation are key for secure blockchain interactions.

Addressing the challenges of adversarial attacks on time series neural networks.

New methods improve machine learning models' defense against adversarial attacks.

This study examines the reliability of rationalization models under adversarial attacks.

Examining Automatic Speech Recognition in Canadian court systems and its impact.

A framework for assessing the security of SDN systems in data centers.