Latest Articles for Data Privacy

Researchers present a cost-effective approach to privacy risks in large language models.

Exploring quantum methods for secure and efficient data retrieval.

Federated Learning offers a new way to protect data while collaborating across devices.

A new method identifies problematic devices in federated learning to improve speed and security.

A look at activation steering for improving AI memory handling.

Learn how hierarchical federated learning protects data while enabling collaboration across devices.

Enhancing realism in synthetic faces for better face recognition performance.

FedPID enhances tumor segmentation while safeguarding patient data.

This article discusses methods for safely sharing cumulative distribution functions.

Research sheds light on tuning hyperparameters for better model performance.

Learn how FedGPL helps organizations collaborate while protecting their data privacy.

This paper discusses creating safe, fair guidelines for AI development.

Research reveals risks of property inference attacks on GNNs and the need for data privacy.

New methods enhance distributed optimization while preserving agent data privacy.

Introducing a method to enhance privacy without sacrificing model accuracy.

Strategies to safeguard images from unauthorized reconstruction.

Exploring new ways to combat money laundering through data sharing and machine learning.

Exploring membership inference attacks to protect data privacy in advanced models.

FedRISE enhances federated learning by filtering bad data updates for better model training.

A new method improves model performance amid varying data distributions.

Researchers develop methods to teach computers to process invoices while protecting privacy.

Federated Learning and Explainable AI ensure safe and clear data practices.

QuanCrypt-FL enhances security in Federated Learning using advanced techniques.

This system enhances DR detection while maintaining patient privacy.

Examining the risks and defenses of Multi-Modal Language Models in AI.

This research highlights innovative methods for privacy in collaborative machine learning.

Exploring how AI can learn while keeping user data private.

New technologies enhance diagnosis and treatment of intracranial hemorrhages.

Federated learning enhances medical imaging while protecting patient privacy.

Find personalized travel suggestions with the INDIANA platform for unique experiences.

A cost-effective way to assess privacy risks in machine learning models.

MaaS offers convenience, but cyber security is crucial for user safety.

Data forging attacks pose serious threats to model integrity and privacy.

A focused approach to rehabilitation that meets individual patient needs for better recovery.

A new approach to ORAM enhances data security and efficiency in memory access.

A new method improves defenses for language models against manipulative attacks.

A method to enhance Federated Learning by adjusting client contributions for better outcomes.

Research shows SNNs may enhance data privacy over traditional models.

A new defense strategy enhances model privacy without sacrificing performance.

A look into secure communication methods, including quantum oblivious transfer.