I am a final-year Ph.D. candidate in Show Lab at National University of Singapore, supervised by Prof. Mike Shou. I received my bachelor’s degree and master’s degree from the Image Processing Center of Beihang University, supervised by Prof. Zhenwei Shi, also working closely with Prof. Zhengxia Zou.

Building omni world models: generative models that can understand, simulate, and interact with the visual world across modalities.

• Intervention: intervening in the virtual world in human desired ways (MotionDirector, P-Flow).
• Autonomy: advancing generative models to evolve and self-evlove (EvolveDirector, DoraCycle).
• Omni-modality: unifying vision and language within a unified modeling framework (DoraCycle, FlowInOne).
• Efficiency & Downstream tasks: making diffusion models faster (Glance), and utilizing the world knowlege in downtream tasks (Dream.exe).

I am open to new ideas, discussions, and possible collaborations. Feel free to reach out.