Qi Cao

MepoGNN combines models to improve predictions of disease spread.

AutoHint improves prompt quality for better language model performance.

Examining how LLMs prefer generated contexts over retrieved ones in answering questions.

Revealing the threats and defenses of recommender systems.