Xiang Li

I am currently a post-doctoral researcher at University of Pennsylvania, working with Prof. Qi Long and Prof. Weijie Su. I obtained my Ph.D. in statistics from School of Mathematical Sciences, Peking University in 2023, advised by Prof. Zhihua Zhang. Before that, I earned a B.S. in Statistics and a B.A. in Economics from Peking University in 2018.

My research interests center around the intersection of statistics, stochastic optimization, and machine learning. During my Ph.D., I worked on federated learning (communication efficiency and data heterogeneity), stochastic approximation (asymptotic behavior and convergence analysis), online decision-making (sample complexity and robustness), and online statistical inference.

More recently, my research has focused on large language models, where I study statistical questions—such as detection, inference, and robustness—while treating the models as black boxes without relying on their internal parameters or architecture. One key area is statistical watermarking, where I develop and analyze methods for embedding and detecting watermarks in generated text. My work aims to enhance the reliability and robustness of these techniques with provable guarantees, with a broader goal of advancing LLM inference and usage.

I am on the job market in the 2025–2026 cycle, mainly considering academic opportunities, with an interest in scientific research roles if pursuing positions in industry.

Contact Info: lx10077 at upenn dot cn