|
Zicheng Liu I am an AI researcher working toward trustworthy, scalable, and physically grounded AI systems. I received my M.Sc. in Artificial Intelligence from the University of Hong Kong and my B.Eng. in Artificial Intelligence from Wuhan University. My goal is to build intelligent systems that can learn from complex data, reason under uncertainty, and interact reliably with the physical world. |
|
ResearchMy current research interests include large language models (LLMs), multimodal large language models (MLLMs), generative AI and embodied AI. |
Selected Publications* denotes equal contribution. |
|
|
“Very Likely” Means “Uncertain”? How LLMs Diverge from Humans in Linguistic Uncertainty Quantification
Jinhao Duan*, Zicheng Liu*, Zijie Liu*, Kaidi Xu, Tianlong Chen International Conference on Machine Learning (ICML), 2026 paper |
|
PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation
Kexu Cheng*, Zicheng Liu*, Mingju Gao*, Chunhe Song, Hao Tang European Conference on Computer Vision (ECCV), 2026 paper / code |
Research Experience | |
| 12/2025 - Present | TH Lab, Peking University Research Intern, advised by Prof. Hao Tang Physical intelligence, Video Generation, and Embodied AI. |
| 06/2025 - 12/2025 | UNITES Lab, UNC-Chapel Hill Research Intern, advised by Prof. Tianlong Chen LLM uncertainty quantification, LLM bias, and Multimodal Reasoning. |
Education | |
| 2024 - 2026 | The University of Hong Kong M.Sc. in Artificial Intelligence, Department of Mathematics. |
| 2020 - 2024 | Wuhan University B.Eng. in Artificial Intelligence, School of Computer Science. |
| Template adapted from Jon Barron's academic website. |