Research

Working Papers

Stress Testing Financial Portfolios with Coverage Guarantees
Zhongze Cai, Martin Haugh, and Xiaocheng Li.

Publications & Preprints

Users as Annotators: LLM Preference Learning from Comparison Mode (arXiv)
Zhongze Cai and Xiaocheng Li, 2025.

Incentivizing High-Quality Human Annotations with Golden Questions (arXiv)
Shang Liu, Zhongze Cai, Hanzhao Wang, Zhongyao Ma, and Xiaocheng Li, 2025.

What Matters in Data for DPO? (arXiv)
Yu Pan, Zhongze Cai, Guanting Chen, Huaiyang Zhong, and Chonghuan Wang, 2025.   Accepted at NeurIPS 2025.

Out-of-distribution Robust Optimization (arXiv)
Zhongze Cai, Hansheng Jiang, and Xiaocheng Li, 2025.   Accepted at UAI 2025.

Towards Better Understanding of In-Context Learning Ability from In-Context Uncertainty Quantification (arXiv)
Shang Liu, Zhongze Cai, Guanting Chen, and Xiaocheng Li, 2024.   Accepted at Transactions on Machine Learning Research.

Towards Better Statistical Understanding of Watermarking LLMs (arXiv)
Zhongze Cai, Shang Liu, Hanzhao Wang, Huaiyang Zhong, and Xiaocheng Li, 2024.   Acceptable after major revisions at Journal of the American Statistical Association.

Distribution-Free Model-Agnostic Regression Calibration via Nonparametric Methods (arXiv)
Shang Liu, Zhongze Cai, and Xiaocheng Li, 2023.   Accepted at NeurIPS 2023.

A Neural Network Based Choice Model for Assortment Optimization (arXiv)
Hanzhao Wang, Zhongze Cai, Xiaocheng Li, and Kalyan Talluri, 2023.   Invited for resubmission at Marketing Science.