I am a Partner Research Scientist at Microsoft's Office of Applied Research.
I obtained my Ph.D. at the University of California, San Diego, advised by Prof. Julian McAuley. Before that, I received my master's in Statistics from UIUC, and my bachelor's in Statistics from Peking University (Yuanpei Program).
My current research explores Intelligence Beyond the Individual: the capabilities that emerge from multiparty interactions among humans, AI agents, or both, and how learning and mechanism design can unlock Collective Intelligence.
On the technical side, my work spans Agentic Reinforcement Learning, LLM Social Reasoning, (Multi-)User Simulation, and RL Environment Design, extending my earlier interests in graph machine learning, network science, and large-scale behavior modeling.
Selected Publications
See Google Scholar for a complete list.
- Teaching Language Models To Gather Information Proactively
- Interactive Speculative Planning: Enhance Agent Efficiency through Co-design of System and User Interface
- TnT-LLM: Text Mining at Scale with Large Language Models
- Interpretable User Satisfaction Estimation for Conversational Systems with Large Language Models
- Workplace Recommendation with Temporal Network Objectives
- Large-Scale Analysis of New Employee Network Dynamics
- Learning Causal Effects on Hypergraphs
Academic Service
I am currently taking a break from paper reviewing, with very few exceptions. I used to actively serve for top-tier conferences and journals in Machine Learning, NLP, and Recommender Systems, and have won several top-reviewer awards.
Workshop Organization
- RecWork 2022 — Recommender Systems for the Future of Work
- RecNLP 2019 — AAAI Workshop on Recommender Systems and NLP
Paper Reviewing
- Conferences AAAI, ACL, AISTATS, CIKM, EMNLP, ICLR, ICML, ICWSM, NeurIPS, RecSys, SDM, TheWebConf
- Journals TPAMI, TKDE, TOIS, TIST, TMLR