Speaker
Yingfei Xiong
Yingfei Xiong obtained his Ph.D. degree from the University of Tokyo in 2009, and worked as a postdoctoral researcher at University of Waterloo from 2009 to 2011. He joined Peking University in 2012 and is currently a tenured associate professor, deputy director of the Institute of Software. His research interest is programming languages and software engineering in general, and program synthesis, repair, analysis and verification in particular. His work has contributed to the development of a series of neural network models for code generation that achieve state-of-the-art performance across different scales, such as the DeepSeek-Coder model; significantly improved the precision, recall, and efficiency of program repair; proposed delta-based BX, one of the two most widely used bidirectional transformation frameworks; successfully solved numerous algorithm design problems, including those from world-class competitive programming contests. His work has been adopted by industry, such as by Huawei, ZTE, and the next-generation Linux kernel configuration project. He has served as the vice chair of OOPSLA, area chair of ASE, and editorial board member of IEEE TSE. He regularly serves on the program committees (PC) of conferences like PLDI, ICSE, FSE, OOPSLA, ASE, and ISSTA, and has received five Distinguished Reviewer Awards at ICSE and FSE. He was awarded the First Prize of the National Technology Invention Award (6th recipient), the First Prize of the Chinese Institute of Electronics Natural Science Award (ranked 1st), the CCF-IEEE CS Young Scientist Award, the MODELS Ten-Year Most Influential Paper Award, and five ACM SIGSOFT/IEEE TCSE Distinguished Paper Awards. He is a Distinguished Member of ACM and a member of IFIP WG 2.4.
Talks at this event:
Sun, 14:00 | Neural Code Generation Models with Programming Language Knowledge |