I am a four-year Ph.D. student in Electrical Computer Engineering at University of Texas at Austin, under the supervision of Prof. Atlas Wang. I received the B.Eng. and M.S. degree (advised by Prof. Li Niu and Prof. Liqing Zhang) in Computer Science and Technology from Shanghai Jiao Tong University, in 2019 and 2022, respectively.

I am currently a research intern at NVIDIA Research Autonomous Vehicle Research Group, mentored by Yan Wang. My current work focuses on reinforcement learning objective design for improving VLA robustness in last-mile autonomous driving scenarios, and released out-of-distribution reasoning benchmarks.

I was a research intern at Adobe Research with Yiqun Mei, at Meta FAIR with Weiyao Wang, Matt Feiszli, at Snap Research with Jian Wang.

I received MLSys'25 Outstanding Paper Award (Honorable Mention) and CVPR'25 AI4CC Workshop Best Paper Award.

My research focuses on modeling dynamic and complex 3D/4D environments: both capturing the real world through digital reconstruction and synthesizing controllable virtual worlds. I am also interested in physical AI for embodied agents, such as autonomous vehicles, especially world-action models and vision-language-action models that couple scene understanding, reasoning, and action generation. I’m also interested in efficient AI algorithms, with an emphasis on improving the training and inference efficiency of large foundation models.