I am a final-year Ph.D. at the University of Pennsylvania under the guidance of Insup Lee and Osbert Bastani. My research is dedicated to enhancing the safety of machine learning, deep learning, and large language models (LLMs). Currently, my work focuses on:
- Uncertainty Quantification
- Safety Alignment
- LLM for Code
- Anomaly Detection
My contributions have been recognized at ML and NLP venues, including NeurIPS, ICLR, ICML, CVPR, NAACL, EMNLP, and KDD. My long-term ambition is to advance the development of safe and secure ML, DL, and LLM systems. I am actively seeking job opportunities and collaborations, and I would be delighted to connect if you’re interested in discussing potential opportunities.
My Motto
“知之为知之，不知为不知，是知也”
“To know what you know and to acknowledge what you do not know—this is true knowledge.”
News
- Our Conformalized credal regions for classification with ambiguous ground truth is accepted to TMLR 2025.
- Our Conformal Structured Prediction is accepted to ICLR 2025.