Latest Articles for Data Security

Research aims to balance privacy and accuracy in vector mean estimation methods.

Two innovative techniques improve adversarial attacks on tabular data models.

Exploring safety, reliability, and ethical issues in language models.

New approach improves machine learning privacy and resilience against attacks.

This method improves data tracking through advanced watermarking techniques.

Fantastyc enhances federated learning through blockchain for improved privacy and security.

A new framework improves healthcare collaboration while protecting patient privacy.

Introducing GENIE, a watermarking method for safeguarding Graph Neural Networks against theft.

New strategies improve deep reinforcement learning against adversarial noise.

A new method enhances targeted attacks using easy samples in neural networks.

This article examines enhancements to SSMs for resilience against adversarial perturbations.

Introducing a method to evaluate model resilience against data poisoning attacks.

Learn how to maintain secure collaboration among parties in cryptography.

A new approach enhances reliability in decentralized communication systems for better security.

This study investigates the limits of accuracy and robustness in machine learning models.

A method combining SGD and privacy techniques for effective data analysis.

A new watermarking method aids in identifying synthetic tabular data.

New methods improve unlearning harmful data in machine learning systems.

A new approach to enhance homomorphic encryption processing speeds.

A new method to verify machine unlearning effectively and securely.

A new method enhances recommendation systems while protecting user privacy.

A new method to safeguard datasets against unauthorized usage in generative models.

A new protocol combines transport and encryption for safe, efficient data movement.

A study on the effectiveness of OOD detectors against adversarial examples.

New methods to protect data privacy in cloud computing environments.

Introducing watermarking to protect counterfactual explanations in machine learning.

A new framework enhances how users query IoT data and classify threats.

A new system facilitates private use of LLMs through HPC resources.

Local Differential Privacy enables secure data sharing while protecting individual user information.

FernUniversität works on open LLM infrastructure to support education.

A new algorithm enhances data exchange among vehicles while ensuring privacy.

Examining the security of SHA-256 and recent collision finding techniques.

New hybrid methods improve efficiency in integer factorization for cryptography.

A novel approach to ensure privacy while maintaining text utility in NLP models.

Quantum communication offers secure data transmission using quantum mechanics principles.

A framework for formal verification of oblivious algorithms to protect sensitive data.

A method for compressing images while ensuring their security during transmission.

A new system enhances data processing while ensuring user privacy and efficient resource use.

Researchers examine methods to secure sensitive information in text classification models.

Exploring backdoor attacks and graph reduction methods in GNNs.