Hongyu Liu

I am a Ph.D. student in the Department of Computer Science and Engineering at The Hong Kong University of Science and Technology [HKUST], where I am fortunate to be advised by [Prof. Qifeng Chen].

My research centers on video generation and generative world models, human-centered multimodal spatial intelligence, and image generation and editing. I study how generative systems can understand, predict, and synthesize visual worlds across space and time, with a particular interest in realistic, controllable, and interactive digital humans and avatars.

I am particularly interested in real-time and long-form video generation, as well as interactive world models that maintain persistent state, spatial consistency, and controllable scene evolution over extended horizons. My goal is to build efficient generative systems that can perceive actions, model future dynamics, and produce coherent, responsive visual experiences for both open-ended worlds and digital humans.

Beyond research, I am deeply passionate about fishing. I hope to one day travel across China to fish in its rivers, lakes, and coastal waters, using the journey to reconnect with nature, explore local cultures, and find ideas outside the lab.