Dr. Xujie Si

Biography

Xujie Si is an Assistant Professor at the University of Toronto in the Department of Computer Science. His research interests focus on improving software quality and advancing state-of-the-art static analysis, verification, synthesis, and testing using AI-based techniques. His work includes applying symbolic reasoning, constraint solving, and statistical and probabilistic models. A significant part of his research involves deep learning and reinforcement learning to address challenges in program reasoning. He emphasizes the importance of automatically learning API specifications from large-scale systems and has developed techniques that combine statistical and logical methods to improve analysis accuracy and reduce false alarms in static analysis tools. His innovative frameworks for scalable logic programming synthesis have had applications in big-data analytics and software-defined networks. Furthermore, he has tackled fundamental challenges in automated software verification, such as finding strong loop invariants and developing end-to-end learning frameworks for program synthesis. His contributions include notable publications in high-impact conferences like NeurIPS, ICLR, and OOPSLA.

Research Interests

Experience