Scholar
Xiang Li
Google Scholar ID: -okA4fgAAAAJ
University of Pennsylvania
Citations & Impact
Citations
3,831
H-index
11
i10-index
15
Publications
20
Co-authors
17
list available
Publications
Resume (English only)
Academic Achievements
- Two papers accepted as Spotlights at NeurIPS 2025: one on empirical evaluation of goodness-of-fit tests for watermark detection, and another on mitigating privacy–utility trade-offs in decentralized federated learning.
- Two papers accepted at AISTATS 2023.
- Paper 'On the convergence of FedAvg on non-iid data' accepted at ICLR 2020 with an oral presentation.
- Recipient of the IMS New Researcher Travel Award (2025).
- Publications in top journals including The Annals of Statistics and Journal of the Royal Statistical Society: Series B.
- Presented work at major conferences including JSM, ICSA, SLDS, MOPTA, Allerton, and NeurIPS.
Background
- Postdoctoral researcher in Statistics at the University of Pennsylvania, working with Prof. Qi Long and Prof. Weijie Su.
- Research interests lie at the intersection of statistics, optimization, and machine learning, with applications in data science and artificial intelligence.
- Current research focuses on the statistical and algorithmic foundations of reliable AI, especially large language models (LLMs).
- Investigates statistical watermarking for provenance and robustness of AI-generated content, and develops tools to evaluate how LLMs encode and use knowledge.
- During PhD, designed methods for learning with heterogeneous and online data, addressing communication efficiency in federated learning, robustness under data heterogeneity, and uncertainty quantification in streaming settings.
