I am a master's student in Artificial Intelligence at the Hangzhou Institute for Advanced Study, University of Chinese Academy of Sciences, advised by Prof. Jungang Xu. I expect to graduate in June 2027. I received my bachelor's degree in Intelligent Science and Technology from Wuhan Institute of Technology in 2024.
My research brings together scientific machine learning, embodied intelligence, and autonomous research agents. I develop physics-aware models for ocean carbon-flux forecasting and study how navigation agents can assess their progress and recover from mistakes. Across these projects, I am interested in systems that use physical structure and feedback to make better predictions and decisions. I also explore agents that plan experiments, interpret results, and iteratively improve scientific models.
I am currently an LLM Algorithm Intern at ByteDance, applying research agents to logistics and supply-chain forecasting for cross-border e-commerce. Previously, I developed an Auto Research framework at Baidu. My earlier work on humanoid robots and a board-game-playing robotic arm shaped my interest in connecting perception, learning, and action—an interest that continues to guide my research today.
Scientific AI
Embodied intelligence
Autonomous research agents
A hierarchical mixture-of-experts model that separates environmental regimes from physical mechanisms for interpretable global ocean carbon-flux forecasting.
Geometrically verified task graphs and state-aligned reinforcement learning help navigation agents assess progress, recognize failure, and recover over long horizons.
A 17-year, approximately 1 km benchmark and a physics-inspired model that pair large-scale transport modeling with iterative refinement of local carbon-flux detail.
Master's student in Artificial Intelligence
Hangzhou Institute for Advanced Study
Advised by Prof. Jungang Xu.
Bachelor's degree in Intelligent Science and Technology
School of Computer Science and Engineering
Admitted to UCAS through the graduate recommendation pathway.
LLM Algorithm Intern
TikTok Global E-commerce · Data
Beijing, ChinaAutonomous research for real-world forecasting
Agent Algorithm Intern
BMU Foundation Model Department · Frontier Research
Beijing, ChinaAgents that plan, explore, and improve
My interest in embodied intelligence began at Wuhan Institute of Technology, where I worked with Tongzhou Zhao and Xun Li.
In the university's robotics laboratory, I worked on humanoid robot vision and control and later led a robotic-arm project for board-game playing using deep learning and reinforcement learning. Those hands-on experiences continue to inform my work on perception, learning, and decision-making.
UNDERGRADUATE ROBOTICS →Vision and control in the university robotics laboratory.
Project lead · Deep learning and reinforcement learning.
My publication list now includes work at ACM MM 2026, IROS 2026, ICASSP 2026, IJCNN 2026, and IEEE Access. Read the papers.
I joined ByteDance as an LLM Algorithm Intern with the TikTok Global E-commerce team.
I joined Baidu to work on autonomous research agents.
