Research focuses on trustworthy AI and software engineering for machine learning, including large language model reliability, automated code refactoring, deep learning testing and bug localization, and verification of model behavior in deployment.