Note
I'm looking for full-time Research Scientist roles starting Summer 2027 (PhD expected Jul 2027). The best way to reach me is email or LinkedIn.
I'm a 4th-year PhD student in Computer Vision at MBZUAI (Abu Dhabi), working in the Metaverse Lab under the supervision of Prof. Hao Li. I work on vision-language models (VLMs) and generative models for video and 3D. Before that, my work on 3D vision (PointNeXt, 3D CoMPaT) was published at NeurIPS and ECCV.
I have done research internships at Amazon Science (Prime Video) and Sony AI, and I serve as a reviewer for CVPR, ICCV, NeurIPS, AAAI, TPAMI, IJCV, TCSVT, and BMVC.
| Year | Paper | Venue | Links |
|---|---|---|---|
| 2026 | Perceptio: Perception Enhanced Vision Language Models via Spatial Token Generation | arXiv:2603.18795 | Paper · HF |
| 2025 | 3DCoMPaT++: An Improved Large-scale 3D Vision Dataset for Compositional Recognition | TPAMI 2025 | Code |
| 2025 | Exploring Scaling Laws of PointNets | 3DV 2025 (Nectar Track, Spotlight) | |
| 2022 | PointNeXt: Revisiting PointNets with Improved Training and Scaling Strategies (1,600+ citations) | NeurIPS 2022 | Code |
| 2022 | 3D CoMPaT: Composition of Materials on Parts of 3D Things | ECCV 2022 (Oral) | Code |
Full list on Google Scholar.
- [Sep 2026] Selected as a Young Researcher at the Heidelberg Laureate Forum (HLF) and awarded the Carl-Zeiss-Stiftung Abbe Grant (top 3%).
- [Mar 2026] Perceptio is out on Hugging Face and arXiv.
- [Sep 2025] 3DCoMPaT++ accepted to TPAMI 2025.
- [Jul 2025] Started as an Applied Scientist Research Intern at Amazon Science (Prime Video), Seattle.
- Amazon Science, Prime Video, Applied Scientist Research Intern (Jul – Nov 2025): scaled distributed training of Perceptio, a perception-enhanced VLM, on 256 NVIDIA H200 GPUs on AWS.
- Sony AI, Research Intern (Jun – Nov 2024): built a music-driven dance video generation pipeline with a diffusion-based video model.
- MBZUAI, PhD in Computer Vision (Aug 2023 – Jul 2027, expected).
- KAUST, M.Sc. in Computer Science (2020 – 2022) · SUSTech, B.Sc. in Computer Science and Technology (2017 – 2021).
- Multimodal Large Language Models and Vision-Language Models
- Video Generation and Video Understanding
- 3D Vision and 3D Scene Understanding
- Generative Models for 3D and Video
- World Models and Embodied AI
Interviews, talks, and podcasts
- 🌸 72小时全纪实：人类与AI的第一次亲密接触 (36Kr, Jun 12, 2025). A 72-hour account of humans' first close encounter with AI.
- 💥 国庆团圆日，海外学子的生活是怎样的？| 「后浪聊天室」Let's chat! (Bilibili, Oct 11, 2022). Life of overseas students on National Day.
- 💰 迪拜商业英才李禹陈：浅谈中东2030愿景——中国创业者在中东创业的机遇与挑战 (Jiacheng Capital). Middle East 2030 vision: opportunities and challenges for Chinese entrepreneurs.
- 📱 07. 去中东上学：全靠酋长资助，不用交学费，还赚钱！ (Xiaoyuzhou podcast). Studying in the Middle East: no tuition, with a stipend.
Website · Google Scholar · LinkedIn · Email