Suhang Wang

A new approach enhances image safety in text-to-image models through prompt optimization.

A new approach enhances graph neural networks using large language models.

Discover how LLKD improves learning efficiency using smaller models.

Learn how calibration improves the accuracy of language models.