Research | Yuxuan Tang 唐玉轩

What I’m most proud of and happiest about is that every project I’ve worked on began from genuine curiosity and interest, not just because it was something I could do. I’m deeply grateful to my advisors for their constant support and understanding.

Overall, I am committed to data-driven, practice-oriented, and interdisciplinary research.

Methodologies: AI/ML, Choice Modeling, Active Learning, Data-Driven Modeling

Applications: LLM Post-training, Preference Learning, E-commerce

Published Papers

On a Mallows-type Model for (Ranked) Choices , with Yifan Feng

Advances in Neural Information Processing Systems, 35, 2022: 3052–3065 (NeurIPS '22)

Working Papers

A Mallows-type Model for Preference Learning from Ranked Choices , with Yifan Feng
Under Second Round Revision at Management Science
- Preliminary version accepted at NeurIPS, 2022
- Featured in EC25 Workshop on New Directions in Social Choice (Spotlight)
- Finalist, POMS-HK Best Student Paper Award Competition, 2025
- Third Prize, CSAMSE Best Paper Award Competition, 2023

Beyond Pairwise: Empowering LLM Alignment With Ranked Choice Modeling , with Yifan Feng

Add to Order in Contingent Free Shipping on Online Grocery Platforms, with Xiaobo Li, Rowan Wang, and Xuan Wang

Work in Progress

Parameterizing the DAG-based Preference with Optimal Pricing Decision, with Xiaobo Li and Shuo Wang
Preference Modeling on Directed Acyclic Graph with the Mallows–RMJ Model.