Xiaoyang Li

I’m an undergraduate student double-majoring in Computer Science and Mathematics at the University of California, Irvine (UCI).

My academic journey at UCI has been shaped by a strong foundation in both theoretical mathematics and practical computing. With a current GPA of 3.965/4.0, I have earned Departmental Honors in Computer Science and am pursuing Honors in Mathematics. My coursework spans advanced topics like stochastic processes, machine learning, and computer vision.

My research interests lie at the intersection of generative models and probabilistic theory, with a particular focus on diffusion models. What draws me to this field is how diffusion elegantly unites stochastic differential equations (SDEs)—which I first encountered in Math 130C—with practical generative AI. Through hands-on projects, I have explored training-free image editing techniques and latent-space video generation on limited hardware, while a separate project on zeroth-order optimization has deepened my appreciation for the mathematical foundations of sampling and black-box optimization.

Looking ahead, I aspire to push the boundaries of diffusion models in areas such as efficient sampling, controllable multimodal generation (e.g., video and 3D), and extensions to discrete spaces like graphs or text. My long-term goal is to pursue a PhD in this area, followed by research in a leading industrial lab (such as Google DeepMind, OpenAI, or Meta FAIR) or academia, contributing to more efficient, interpretable, and versatile generative systems.

Outside of research, I enjoy tackling algorithmic problems, experimenting with new ideas in code, and mentoring peers as a learning assistant. This blog will document my thoughts on generative AI, probabilistic modeling, research experiences, and the occasional mathematical insight.