Jiajun Bao

I am a third-year PhD student in the Center for Applied Mathematics at Cornell University, advised by Prof. Christopher Earls.

My research lies at the intersection of applied mathematics and machine learning. Broadly, I am interested in using data-driven methods to model, predict, and control the dynamics of complex systems.

Currently, I focus on large language model interpretability, studying LLMs as complex systems through the lens of dynamical systems theory and graph signal processing. I investigate how these models learn, represent, and reason from data. My goal is to use interpretability, grounded in a mathematical perspective, to understand, learn from, and guide the design of AI systems.

Prior to Cornell, I received my M.S. in Computational and Applied Mathematics from the University of Chicago, where I was advised by Prof. Daniel Sanz-Alonso, and my B.S. in Applied Mathematics from the University of Washington, where I worked with Prof. Jason Bramburger.