I am an undergraduate student at the School of Intelligent Science and Technology, Nanjing University, advised by Prof. Ying Tai. I will join NJU as a M.S. student in Fall 2026.
My research interests lie in Text-to-Image Generation, Text-to-Video Generation, and Agentic AI. I am passionate about building powerful and controllable generative models that bridge language and visual content, as well as developing autonomous agents capable of complex reasoning and decision-making.
I was awarded the National Scholarship (Top 0.2% nationwide) and ranked in the Top 10% of my school for multiple academic years.
🔥 News
- 2026.01: 🎉🎉 Our paper VGGT-Long has been accepted by ICRA 2026!
- 2025.09: 🎉🎉 I have been admitted to the M.S. program at the School of Intelligent Science and Technology, Nanjing University!
- 2024.12: 🎉🎉 I have been awarded the National Scholarship (Top 0.2% nationwide, the most prestigious undergraduate scholarship in China)!
- 2024.5: 🎉🎉 Meritorious Winner (M Award), Mathematical Contest in Modeling (MCM), COMAP (Top 7% worldwide)
📝 Publications
VGGT-Long: Chunk it, Loop it, Align it — Pushing VGGT’s Limits on Kilometer-scale Long RGB Sequences
Kai Deng, Zexin Ti, Jiawei Xu, Jian Yang, Jin Xie
- A training-free system extending VGGT to kilometer-scale outdoor 3D reconstruction via chunk-based processing, overlapping alignment, and loop closure optimization.
🎖 Honors and Awards
- 2025.11 People’s Scholarship (First Class), Nanjing University
- 2024.12 National Scholarship, Ministry of Education of China
- 2024.7 Third Prize, National College Student Service Outsourcing Innovation Competition
- 2024.5 Meritorious Winner (M Award), Mathematical Contest in Modeling (MCM), COMAP
- 2023.11 People’s Scholarship (Third Class), Nanjing University
📖 Educations
- 2026.09 - present, M.S. in Intelligent Science and Technology, Nanjing University, advised by Prof. Ying Tai
- 2022.09 - 2026.06(expected), B.S. in Intelligent Science and Technology, Nanjing University
💻 Internships
- 2025.02 - 2025.06, Research Intern, EPIC Lab, Shanghai Jiao Tong University, China. Focused on video data distillation and compression.