Xinlei He

A method to forecast non-factual answers from language models before they generate responses.

This study examines watermarking methods for machine-generated text and their effectiveness against removal attacks.

Examine various jailbreak attacks on language models and their defenses.

Exploring privacy risks in masked image modeling and their implications.

ADC streamlines dataset creation, improving speed and accuracy in machine learning.

Research highlights methods to detect backdoor attacks in fine-tuning language models.

Innovative methods aim to enhance AI language models while ensuring safety and efficiency.

Discover how federated learning protects your data while enhancing technology.

Discover how backdoor attacks challenge the safety of AI-driven language models.