Latest Articles for Data Privacy

CDMs help protect sensitive training data while generating high-quality images.

A new protocol enhances time-lock puzzles for secure future communications.

Automated attacks on DNN accelerators can reveal model details with minimal effort.

A new method improves privacy and model training in federated learning.

Learn about the risks of sound-based keyboard attacks and how to protect yourself.

CABB enhances model adaptation while protecting sensitive information across industries.

Learn about gradual sensitivity typing and its role in enhancing programming security.

Examining the potential and challenges of Federated Machine Learning in preserving privacy.

New framework improves QNN training with encrypted data and privacy protection.

A new method enhances federated learning security against data poisoning attacks.

New method reveals vulnerabilities in Vertical Federated Learning systems using Graph Neural Networks.

A method for better image generation that safeguards privacy using public datasets.

PPIMCE enhances performance in secure data processing without compromising privacy.

Exploring the use of synthetic health data to improve CDSS development.

Federated learning allows secure collaboration while keeping sensitive data private.

A new method for secure data aggregation aims to boost efficiency and privacy.

A new chiplet design boosts the efficiency of Fully Homomorphic Encryption.

A new method improves meeting summaries while protecting sensitive data.

A new method for adapting models to real-world data while preserving privacy.

Examining the use of link decorations in online tracking.

Federated Learning enables data privacy while improving machine learning collaboration among diverse participants.

New tool aims to protect analysts and speed up abuse report processing.

Learn how blind annotation enhances data privacy during entity matching.

A new method for lowering expenses in machine learning language services.

A new approach to enhance data privacy while minimizing communication costs.

Explore an innovative defense method to enhance federated learning security.

New methods enhance speed and efficiency of language models on local devices.

New methods to secure federated learning against attacks while preserving user privacy.

A toolchain for enhancing data usage visibility among employees.

Addressing privacy risks while providing valuable insights from machine learning models.

A novel method analyzes daily activities while protecting user privacy.

Creating diverse images of children's faces to enhance recognition systems.

A new method improves segmentation of organs and tumors in medical images.

AIGC redefines how we generate digital content efficiently.

IDiff-Face method improves synthetic data for better face recognition.

Analyzing how step count data can expose personal information and privacy risks.

Exploring the complexities and solutions of machine unlearning for user data privacy.

This research focuses on improving efficiency and privacy in federated learning through adaptive compression methods.

Synthetic data sharing improves health research while protecting privacy.

Combining DNNs and SNNs with homomorphic encryption to ensure data privacy.