Hi!
Welcome! I am an Applied Scientist on the Accio team at Alibaba, working on agentic post-training. Previously, I interned at Amazon, Meta, and IQVIA.
I received my PhD in Statistics from the University of California, Irvine, advised by Prof. Hengrui Cai. I previously studied Biostatistics as a master's student at the University of Washington and received my BSc in Applied Mathematics from Xi’an Jiaotong-Liverpool University.
Research Interests
My research focuses on developing capable and trustworthy language models and agents. I am particularly interested in:
- Agentic post-training, with a focus on long-horizon reinforcement learning, supervised fine-tuning, and on-policy distillation.
- Evaluation: reliable and cost-efficient evaluation of LLM agents.
- Trustworthy AI: uncertainty quantification and interpretability.
Selected Publications
See all publications or my Google Scholar profile for the full list. (* equal contribution)
-
-
OTROPE: Optimal Transport-based Robust Off-policy Evaluation for Large Language Models
-
Towards Bridging the Gap Between Offline and Iterative Alignment via Preference Distillation
-
Reasoning Is Not Free: Robust Adaptive Cost-Efficient Routing for LLM-as-a-Judge
-
Where to Intervene: Action Selection in Deep Reinforcement Learning
-
Interpretable discriminant analysis for functional data supported on random nonlinear domains with an application to Alzheimer’s disease
-
Towards trustworthy explanation: On causal rationalization
Industry Experience
Alibaba
Applied Scientist · Accio
Academic Service
- Area Chair, ACL Rolling Review (ARR), October 2026.
- Conference Reviewer: ICML, NeurIPS, ICLR, AISTATS.
- Journal Reviewer: TMLR, JASA.
Education
University of California Irvine
PhD in Statistics
University of Washington
Master in Biostatistics
Xi’an Jiaotong-Liverpool University
Bachelor in Applied Mathematics