Scholar
Zhengxiong Luo
Google Scholar ID: Sz1yTZsAAAAJ
Bytedance Seed
Citations & Impact
Citations
2,450
H-index
14
i10-index
17
Publications
20
Co-authors
21
list available
Contact
Publications
Resume (English only)
Academic Achievements
- Selected publications:
- - Mogao: An omni foundation model for interleaved multi-modal generation
- - Emu3: Next-Token Prediction is All You Need
- - End-to-End Alternating Optimization for Real-World Blind Super Resolution
- - VideoFusion: Decomposed Diffusion Models for High-Quality Video Generation
- - Learning the Degradation Distribution for Deep Blind Super Resolution
- - Efficient Human Pose Estimation by Learning Deeply Aggregated Representations
- - Rethinking the Heatmap Regression for Bottom-up Human Pose Estimation
- - Unfolding the Alternating Optimization for Blind Super Resolution
Research Experience
- Worked as a full-time researcher at the Beijing Academy of Artificial Intelligence (BAAI) from July 2023 to February 2025; currently a full-time researcher at ByteDance Seed.
Education
- Bachelor's degree in Mechanical Engineering from Shanghai Jiao Tong University (SJTU); Ph.D. from the Institute of Automation, Chinese Academy of Sciences (CASIA).
Background
- Current research focuses on unifying multimodal generation and understanding. Previous work has covered human pose estimation, low-level vision tasks, and video generation.
