Biography
I’m a Ph.D. student at CUHK, advised by Jiaya Jia and Bei Yu. I’m currently a Research Scientist Intern at Meta Superintelligence Labs, working with the Segment Anything Model (SAM) Team.
My research is in multimodal AI and robotics. Previously, I focused especially on efficient vision-language models, visual reasoning, and vision-language-action systems. (The AI community moves fast lol.) I always hope to stay at the frontier of AI and help push it forward.
News
Experience
Research Scientist Intern
- Core contributor to Mage-VL, a codec-native streaming vision-language model.
- Designed for proactive visual perception.
Research Scientist Intern
- Research resulted in a NeurIPS 2025 publication.
- Conducted research on multimodal foundation models, with a focus on efficient visual representation and vision-language reasoning.
Research Assistant
- Collaborated with Prof. Shanghang Zhang.
- Conducted research on domain adaptation and robotics.
Research Assistant
- Collaborated with Prof. Stan Z. Li.
- Conducted research on contrastive learning and manifold learning.
Awards
- NeurIPS Top Reviewer
- ICLR Notable Reviewer
- SenseTime Scholarship
- 1st Place - ICCV2023 workshop competition - Visual Continual Learning
- Silver Prize (top 3%) - ECCV2022 workshop competition - Universal Image Embedding Challenge
- Silver Prize (top 2%) - Google Brain - Ventilator Pressure Prediction
Contact
- yangsenqiao.ai@gmail.com
- +852 68749085
- Shatin, New Territories, Hong Kong