I am a Ph.D. candidate in Computer Science at the University of Southern California, advised by Prof. Xuezhe Ma. Before joining USC, I earned an M.S. in Computer Science from New York University, where I worked with Prof. Yi-Jen Chiang, Prof. Christopher Musco, and Prof. Yi Fang. Previously, I studied Electrical Engineering as an undergraduate at the University of Michigan, Ann Arbor.
My current research focuses on optimizers for large-scale language model pre-training. I study the mathematical principles underlying training dynamics across model scales, with related work on hyperparameter transfer and scaling-aware parameterizations such as μP. I am also interested in efficient architectures for sequence modeling, and the interplay between architecture and optimization. My earlier work spans transfer learning, 3D learning, and streaming algorithms.
For the full publication list, please refer to Google Scholar. (* indicates equal contribution.)
Alongside my Ph.D., I am enrolled in a master’s program in mathematics in USC’s Department of Mathematics. When I am not doing research, math study takes a majority of time. One topic I really like comes from Math 525b. It gives a flavor of functional analysis. I wrote and maintain a set of notes:
Notes on Functional Analysis (updated 6/27/2026)
It’s a bit long, but I hope it could be helpful in some way. Feel free to let me know if you spot any errors. I passed my Math Qualifying Exam in Probability, and wish me luck on the Analysis and Algebra exams!
|
Website template. Last updated: November 8, 2026. |