About

Hi! 👋 I am Yihang Du (杜艺航). I am an incoming PhD student at The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), where I am advised by Prof. Benyou Wang at FreedomAI Lab. I am also affiliated with the Shenzhen Loop Area Institute (SLAI) and collaborate with industry research teams such as the VLA team@Lion Rock AI Laboratory.

My research focuses on foundation models for perception, generation, and interaction, with a particular interest in world models and native multimodal models. I aim to understand how models learn generalizable representations of the world, how such representations support generation, reasoning, and interaction, and how we can build more capable and efficient foundation models.

Currently, my research interests include:

• World Models: Learning scalable generative world simulators through autoregressive video generation, post-training, and interactive environment modeling.
• Native Multimodal Models: Exploring unified architectures that jointly learn multimodal understanding and generation.
• Representation Learning for Foundation Models: Studying how different modeling paradigms (e.g., autoregressive, diffusion-based, and unified multimodal architectures) acquire and organize internal representations, and how these representations enable generalization.

I am actively seeking internship and research collaboration opportunities in Multimodal AI. Please feel free to reach out if you are interested in collaborating.

Education

Experience

Misc

I am a big fan of Podcasts and I enjoy playing table tennis. I am trying to learn tennis.
I love photography. I capture the beauty of my travels with a SONY A7M4. Here are some of my photos.