Zhibo Zhu

AI4GC Lab

Zhibo Zhu

Video GenerationDiffusion ModelsVDM Inference Acceleration

About Me

Hi! I'm Zhibo Zhu, an incoming master's student at AI4GC Lab, Zhejiang University, advised by Prof. Shengyu Zhang. I will join the lab in September 2026.

My current research centers on efficient video generation, especially making video diffusion models faster and more practical at inference time.

Research Interests

  • Video Generation - Developing efficient video generation methods that balance visual fidelity, temporal consistency, and computational cost.
  • Diffusion Models - Studying diffusion-based generative models, especially the feature dynamics that emerge during iterative denoising.
  • VDM Inference Acceleration - Accelerating video diffusion models through training-free caching, feature reuse, adaptive block skipping, and test-time self-calibration.

Education

  • 2026 - present · M.S. in Software Engineering, Zhejiang University
  • 2022 - 2026 · B.S. in Computer Science and Technology, Minzu University of China