Doctoral research in trustworthy AI focusing on scalable uncertainty quantification for large language models, developing Bayesian and probabilistic methods to align model confidence with user perception and enhance NLP system reliability.