01
Video Generation
Generating realistic videos with stronger physical consistency and precise, flexible control.
Physical Controllable
Hello, I'm
Hang Liu
I am a fourth-year undergraduate in Artificial Intelligence at Xiamen University. In Fall 2027, I will join Peking University as a graduate student, advised by Prof. Li Yuan.
I am fortunate to work as a research intern at Xiamen University MAC Lab, under the supervision of Prof. Jiayi Ji and Prof. Rongrong Ji. Previously, I worked as a research intern with Prof. Di Hu at the Gaoling School of Artificial Intelligence, Renmin University of China.
01 / Updates
Latest News
Our paper MICON-Bench was accepted to CVPR 2026.
Awarded the National Scholarship.
Our paper RollingQ was accepted to ICML 2025.
Research internship at DeepRoute.ai, working on model R&D and optimization for autonomous driving.
Joined Xiamen University MAC Lab as a research intern.
Joined Gewu Lab at the Gaoling School of Artificial Intelligence, Renmin University of China, as a research intern.
02 / Research
Research Interests
My research lies at the intersection of multimodal learning and generative models. I am interested in how understanding and generation can reinforce each other to produce more general, robust, and physically grounded intelligence.
02
World Models
Learning predictive representations of the world for simulation, planning, and embodied agents.
03
MLLMs & UMMs
Exploring multimodal understanding and reasoning, together with unified models that connect perception and generation.
03 / Selected Work
Publications
CVPR 2026 Co-first Author
MICON-Bench: Benchmarking and Enhancing Multi-Image Context Image Generation
Mingrui Wu*, Hang Liu*, Jiayi Ji, Xiaoshuai Sun, Rongrong Ji
ICML 2025 Conference Paper
RollingQ: Reviving the Cooperation Dynamics in Multimodal Transformer
Haotian Ni, Yake Wei, Hang Liu, Gong Chen, Chong Peng, Hao Lin, Di Hu
Let's connect