About

I’m a second year PhD student in Machine Learning at Georgia Tech. I work on making large language models (LLMs) more efficient to train and use, with recent focus on optimization algorithms, synthetic data for model post-training, and diffusion language models (DLLMs). I’ve collaborated closely with Amazon Rufus and Microsoft GenAI, and I enjoy building practical methods that improve both model performance and compute efficiency. Before grad school, I studied Statistics and Economics at Peking University.