Yihan Wang

Analyzing stability in adversarial training to enhance model generalization.

Understanding unlearnable example attacks through game theory for better data protection.

New methods to safeguard sensitive data against unauthorized access in machine learning.

A new approach using backtranslation aims to protect language models from harmful prompts.

Introducing new methods to improve forgetting processes in contrastive learning models.

Research shows ways to enhance context awareness in language models for better responses.

Learn about data protection methods and threats in the machine learning landscape.