top of page

​​

Foundation Models and LLMs

[Mechanism Design for LLMs, Uncertainty Quantification with ML Models]

(Note:     students under my supervision;   equal contribution;  * corresponding author)​​​​​

​

Mechanism Design for LLMs​​

(Overview)​

  • Mechanism design meets large language models: Foundations and frontiers. [pdf][code​]

Baiting Chen, Tong Zhu, Xuanang Li, Yichi Zhang, and Xiaowu Dai*.

Preprint, 2026.​

 

(Hallucination)​

Baiting Chen†, Tong Zhu†, Jiale Han, Lexin Li, Gang Li, and Xiaowu Dai*.

Advances in Neural Information Processing Systems (NeurIPS), 2025.​

​​

(Reasoning)​​

  • ALIGN: Aligned delegation with performance guarantees for multi-agent LLM reasoning. [pdf​][code​]

Tong Zhu†, Baiting Chen†, Jin Zhou, Hua Zhou, Sriram Sankararaman, and Xiaowu Dai*.

Preprint, 2026.

​​​

(Alignment)

  • Common-agency games for multi-objective test-time alignment. [pdf​][code]

Baiting Chen†, Tong Zhu†, Rui Yu†, and Xiaowu Dai*.

Preprint, 2026.

​

​

Uncertainty Quantification with ML Models​​​

(Prediction-Powered Inference)

  • Surrogate-powered inference: Regularization and adaptivity. [pdf​]

Jianmin Chen, Huiyuan Wang, Thomas Lumley, Xiaowu Dai, and Yong Chen.

Under Moderate Revisions at Journal of the American Statistical Association (JASA), 2025.

​​

  • Prediction-powered conditional inference. [pdf​][code]

Yang Sui, Jin Zhou, Hua Zhou, and Xiaowu Dai*.

Preprint, 2026.

​​​

(LLM Evaluation)

  • Prompt-dependent ranking of large language models with uncertainty quantification. [pdf​][code​]

Angel Rodrigo Avelar Menendez, Yufeng Liu, and Xiaowu Dai*.

Preprint, 2026.

​​

  • Sweeping failures out of the pooled rug: Simultaneous PPI for weakest-stratum confidence in routed LLM judges. [pdf​]

Jiahang Sha, Xiaowu Dai, and Wei Wang.

Preprint, 2026.

​

​

​

​

​

​

nsf-logo-298x300.png
NIH-logo_150.png
ucla.png
logo_hellman_fellows_fund_200.png
amazon.jpeg
merck.png
© 2026 Xiaowu Dai
bottom of page