About

I’m Tianze Xu (“徐天泽” in Chinese), a second-year Ph.D. student at Shanghai Jiao Tong University, advised by Prof. Pengfei Liu. My research interests currently focus on fine-grained post-training (Token-level reward assignment, On-Policy Distillation, etc.) and LLM evaluation.

Previously, I interned at Alibaba (Taobao & Tmall Group) on agentic reinforcement learning, mentored by Yanzhao Zheng.

News

Sep 2026: MOPD-Router: Rethinking Teacher Routing in Multi-Teacher On-Policy Distillation released on arXiv. Code is available.
Jul 2026: 🎉🎉 Four papers have been accepted by COLM 2026!
Apr 2026: Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks released on arXiv.
Jul 2025: ResearcherBench: Evaluating Deep AI Research Systems on the Frontiers of Scientific Inquiry (co-first author) released on arXiv.
Oct 2024: Started my Ph.D. at Shanghai Jiao Tong University, advised by Prof. Pengfei Liu.

Publications

MOPD-Router: Rethinking Teacher Routing in Multi-Teacher On-Policy Distillation Tianze Xu, Yanzhao Zheng, Zhentao Zhang, …, Gang Yu arXiv preprint, 2026

ResearcherBench: Evaluating Deep AI Research Systems on the Frontiers of Scientific Inquiry Tianze Xu, Pengrui Lu, Lyumanshan Ye, Xiangkun Hu, Pengfei Liu Conference on Language Modeling (COLM), 2026

Rubrics to Tokens: Bridging Response-level Rubrics and Token-level Rewards in Instruction Following Tasks Tianze Xu, Yanzhao Zheng, Pengrui Lu, …, Gang Yu Conference on Language Modeling (COLM), 2026

Education

Ph.D. in Electronic Information (Artificial Intelligence), School of Computer Science and Engineering, Shanghai Jiao Tong University, 2025 – 2030 (expected)
B.S., Beijing University of Posts and Telecommunications, 2021 – 2025