Tian Ye
Scholar

Tian Ye

Google Scholar ID: 1sGXZ-wAAAAJ
The Hong Kong University of Science and Technology (Guangzhou)
Computer VisionGenerative ModelFoundation Models
Citations & Impact
All-time
Citations
2,409
 
H-index
23
 
i10-index
34
 
Publications
20
 
Co-authors
4
list available
Resume (English only)
Academic Achievements
  • Publications: Multiple papers accepted by top venues including CVPR 2025 (3), ICCV 2025 (3), ECCV 2024 (2), and IEEE TPAMI 2025 (MovieChat+)
  • Model Releases: Led or contributed to key models such as LucidFlux-14B (SOTA universal image restoration DiT), Flux.1-lite-8B-GRPO (RL post-trained high-quality model), PosterCraft (aesthetic poster generation), MagicInfinite (infinite talking video generation), Magic 1-For-1 (4-step image-to-video diffusion model), and Meissonic-1B (high-res non-AR T2I model)
  • Awards: KAUST AI Rising Star 2025, Outstanding Reviewer at BMVC 2024, ICLR 2025 Notable Reviewer, HKUST(GZ) PG Scholarship 2024, 3rd Place in CVPR 2022 NAS Competition (Supernet Track), JMU Student Star Award 2022 (top 20 out of 19,000)
  • Academic Service: Reviewer for major conferences (AAAI, CVPR, ICCV, ECCV, NeurIPS, ICLR, ICML, etc.) and journals (IJCV, IEEE TIP, etc.); Organizer of LOVEU@CVPR 2024 workshop competition
  • Mentoring: Advising MPhil student Fei Song at HKUST(GZ)