01
About
I'm a graduate student in Computer Science & Technology at Fudan University (2025β2028), advised by Prof. Siyu Zhu in the Fudan Generative Vision Lab.
My research applies generative models to autonomous driving β masked-diffusion vision-language-action (VLA) frameworks and discrete flow matching for motion planning. I'm interested in making driving policies more controllable, efficient, and reliable, and in building open-source systems that turn these ideas into things people can run.
Autonomous Driving
Vision-Language-Action
Diffusion Models
Flow Matching
Motion Planning
03
Projects
Autonomous Driving
WAM-Diff
A masked-diffusion VLA framework with mixture-of-experts and online reinforcement learning for end-to-end autonomous driving.
PyTorchDiffusion VLAMoE + RL
View repository β
Autonomous Driving
WAM-Flow
Parallel coarse-to-fine motion planning via discrete flow matching, generating accurate driving trajectories in few inference steps.
PyTorchFlow MatchingPlanning
View repository β
04
Education
2025 β 2028
Fudan University Β· College of Computer Science and Artificial Intelligence
M.S. in Computer Science & Technology Β· advised by Prof. Siyu Zhu
2021 β 2025
Jilin University Β· College of Software
B.Eng. in Software Engineering