Published papers such as 'Hyperbolic Fine-tuning for Large Language Models' (NeurIPS 2025), 'Towards Non-Euclidean Foundation Models: Advancing AI Beyond Euclidean Frameworks' (WWW 2025). Three NeurIPS 2025 papers accepted, one of which was selected as a spotlight presentation.
Research Experience
Currently an assistant professor at HKUST(GZ) and affiliated with The Hong Kong University of Science and Technology. Previously, a Postdoc Researcher at Yale University.
Education
PhD from The Chinese University of Hong Kong; Postdoc Research at Yale University.
Background
Currently an assistant professor at the Thrust of Artificial Intelligence at Hong Kong University of Science and Technology (Guangzhou), with a research focus on exploring the connection and alignment between the learning space, especially the hyperbolic space, and the underlying geometries, structures, and patterns in different types of data. Recent research topics include: Multi-modal LLMs & LLMs(Peft, RAG, Agent, Reasoning) and non-Euclidean LLMs, Recommender Systems, Personalization, Knowledge graph, and Network sciences, Hyperbolic Representation Learning, Differential Geometry, Hierarchical Modeling, Graph Learning and AI4SCI, Applications of LLMs and Multimodal LLMs in PCB Layout Routing and Personalized AI Copilot.
Miscellany
Looking for self-motivated (visiting) PhD students / (remote) research assistants to work on the above-mentioned topics. Provides GPUs and salaries to qualified candidates. Shares new papers, blogs, and books on hyperbolic representation and deep learning through a Slack channel and GitHub repository.