Yixuan Zhu

I am a fourth-year Ph.D. student at Tsinghua University, advised by Prof. Jie Zhou, Prof. Jiwen Lu, and Prof. Yansong Tang.

Prior to my doctoral journey, I received my B.S. degree in Electronic Engineering with honours from Tsinghua University in 2022.

My academic pursuits revolve around the dynamic intersection of multimodal world models and unified multimodal understanding and generation models. This compelling research area fuels my passion for exploring innovative solutions and contributing to the cutting-edge advancements in the field.

Email / Github

(*Equal Contribution, #Corresponding Author)

We introduce Astra, an interactive world model that delivers realistic long-horizon video rollouts under a wide range of scenarios and action inputs.

We introduce VARestorer, a one-step VAR distillation framework for real-world image super-resolution that mitigates error accumulation.

We introduce FADE—a training-free yet highly effective video editing approach that fully leverages the inherent priors from pre-trained video diffusion models via frequency-aware factorization.

We propose InstaRevive, a straightforward yet powerful image enhancement framework that employs score-based diffusion distillation to harness potent generative capability and minimize the sampling steps.

We proposed a unified framework for various efficient image enhancement tasks with generative diffusion priors.

We propose a new method to exploit diffusion priors for human mesh recovery (HMR) in occlusion and crowded scenarios.

Website Template