About me

I am an Assistant Professor at the Department of Electronic Engineering at Tsinghua University. My research explores the interdisciplinary realms of Artificial Intelligence, Data Science, Social Computing and Urban Science, aiming to develop scientific methods and algorithmic tools to address the long-standing puzzles in the complex networks arising from human dynamics and social interactions. Currently, I am particularly interested in the emergence of role-playing and commonsense reasoning capabilities in Large Language Models (LLMs). My latest research focuses on creating autonomous LLM agents, which are capable of learning from the interactions with the environment, simulating realistic human behavior, and revealing the collective dynamics of societies.

Before my current position, I was a postdoctoral scholar at the University of Chicago. I obtained my Ph.D. in Electronic Engineering from Tsinghua University in 2020.

Prospective Students

Our lab has several open research positions for prospective graduate students (Ph.D./Master programs) and undergraduate research interns. We are looking for self-motivated students to work together on various exciting data science projects. If you are interested, please drop me an email with your CV.

News

• [01/2025] Paper accepted to WWW’25: Division-of-Thoughts: Harnessing Hybrid Language Model Synergy for Efficient On-Device Agents.
• [01/2025] Paper accepted to ICLR’25: Agentsquare: Automatic LLM Agent Search in Modular Design Space. Welcome to try out our open platform!
• [01/2025] We are organizing the AgentSociety Challenge at WWW’25!
• [10/2024] Paper published in Nature Communications: Deep learning resilience inference for complex networked systems.
• [09/2024] Paper accepted to NeurIPS’24: HLM-Cite: Hybrid LanguageModel Workflow for Text-based Scientific citation Prediction.
• [09/2024] Paper accepted to Nature Human Behaviour: Mobility Behaviour Data Quantifies Experienced Inequalities in Urban Space.
• [08/2024] Paper published in KDD’24: Large Language Model-driven Meta-structure Discovery in Heterogeneous Information Network.
• [08/2024] We organized the second Research Summit of Urban Science.
• [08/2024] New preprint on LLM-enhanced intent-aware mobility prediction.
• [08/2024] New preprint proposing a “Perceive, Reflect, and Plan” LLM agentic workflow for goal-directed city navigation.
• [05/2024] We organized a tutorial at TheWebConf’24: Simulating Human Society with Large Language Model Agents. For more details, please refer to our survey paper.
• [02/2024] New preprint introducing DefInt (short for Default-Interventionist), an efficient LLM reasoning framework that exploits the synergistic effect between large and smaller language models.
• [02/2024] New preprint on leveraging the reasoning power of LLM to generate realistic human mobility data.
• [01/2024] Paper published in Nature Humanities and Social Sciences Communications: Counterfactual mobility network embedding reveals prevalent accessibility gaps in US cities.
• [23/2023] New preprint on developing LLM agents in simulated urban environments.
• [08/2023] We organized the first Research Summit of Urban Science.
• [04/2023] I joined the EE Department @ Tsinghua University as an Assistant Professor.
• [08/2022] Paper published in Nature Human Behaviour: Strategic COVID-19 vaccine distribution can simultaneously elevate social utility and equity.
• [06/2022] Paper published in PNAS: Flat teams drive scientific innovation.
• [12/2021] Paper published in Nature Computational Science: Emergence of urban growth patterns from human mobility behavior.