Qingtao Yu (Terry) · 于庆涛
PhD Student · Australian National University
I am a third-year PhD student at the Australian National University, supervised by Dylan Campbell, Hongdong Li, and Tongliang Liu. I expect to graduate in 2027.
My research focuses on generative models for computer vision, with a particular interest in enhancing and adapting foundation models at inference time for diverse downstream applications.
I grew up in Dalian, China.
📌 Canberra, Australia 🎂 2000.02.26
📮 Terry.Yu@anu.edu.au 📱 +61 433665842 💬 WeChat: yqt2000226
If you think I could be a good fit, please don't hesitate to contact me at Terry.Yu@anu.edu.au.
🎓 Education
🏅 Recipient of Full Scholarship
🏅 First-Class Honours, GPA 6.59/7
🚀 Lead Research Projects
TokenRepel: Generating Diverse Image Sets Without Compromising Quality
Probability Density Geodesics in Image Diffusion Latent Space
When 3D Bounding-Box Meets SAM: Point Cloud Instance Segmentation With Weak-and-Noisy Supervision
Offline Makes Online Better: Calibrated Cache Compression for Video Generation
Rethinking Test Time Scaling for Flow-Matching Generative Models
👥 Collaborative Research Projects
Disen-Forcing: Disentangling Semantic Anchoring from Motion for Autoregressive Video Diffusion
Contributed to idea discussion.
Benchmarking Audio Visual Segmentation for Long-untrimmed Videos
Developed an efficient video annotation UI using foundation models.
AutoRefiner: Improving Autoregressive Video Diffusion Models via Reflective Refinement Over the Stochastic Sampling Path
Contributed to idea discussions on noise refinement strategies for image generative models.
💼 Work Experience
Tutored Computer Vision, Computer Graphics, and Statistical Machine Learning
🌟 Professional Service
NeurIPS 2026 · ICLR 2026 · AJCAI 2025 · DICTA 2026 · Robotics and Automation Letters (RA-L)
AIESEC Global Volunteer: taught at a primary school in Azerbaijan
Elected International Students Representative, Ursula Hall, ANU