Focus on the research of artificial intelligence and its applications
I am a doctoral student majoring in Software Engineering at University College London and am currently engaged in research related to deep learning and software engineering. My research interests mainly focus on fields such as computer vision, natural language processing, and the application of AI Agents.
In the past few years, I have focused on developing new deep learning algorithms to solve practical problems, particularly making progress in multi-modal generation. I believe that artificial intelligence technology can bring positive changes to society and am committed to combining theoretical research with practical applications.
Research Interests
- Computer Vision and Image Processing
- Natural Language Processing and Text Mining
- Multimodal learning and cross-domain transfer
Education and Work Experience
- 2025-present - University College London, Software Engineering, PhD Student
- 2024-2025 - Southern University of Science and Technology, Statistics, Research Assistant
- 2022-2024 - Li Auto Inc., Algorithm Engineer
- 2019-2022 - Guangxi University, Computer Science and Technology, Master's Degree
- 2014-2018 - Zhengzhou University, Biology Engineering, Bachelor's Degree
Publications
-
MultiModal Large Language Model with RAG Strategies in Soccer Commentary Generation.
Xiang Li, Shuaishuai Zu, Kevin Zhang, et al.
IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 2025 -
SCBench: A Sports Commentary Benchmark for Video LLMs.
Kuangzhi Ge, Lingjun Chen, Kevin Zhang, Yulin Luo, Tianyu Shi, Liaoyuan Fan, Xiang Li, Guanqun Wang, Shanghang Zhang
-
MCRE: Multimodal Conditional Representation and Editing for Text Motion Generation.
Tengjiao Sun, Xiang Li, Tianyu Shi, et al.
European Conference on Computer Vision (ECCV) Workshops 2024: Foundation Models for 3D Humans -
Uniform Text-Motion Generation and Editing via Diffusion Model.
Ruoyu Wang, Xiang Li(Co-first author), Tengjiao Sun, et al.
NeurIPS 2024 Workshop on Adaptive Foundation Models -
Developing a Deep Learning Network MSCP-Net to Automatically and Accurately Generate Maize Stalk Anatomical Traits Related with Plant Lodging Resistance and Yield.
Haiyu Zhou, Xiang Li (Co-first author), Yufeng Jiang, et al.
European Journal of Agronomy. In Press