Jiajun Liang
Scholar

Jiajun Liang

Google Scholar ID: xNxlvjEAAAAJ
MEGVII, KLINGAI
Deep LearningComputer VisionVision-Language LearningGenerative Model
Google Scholar↗
Citations & Impact
All-time
Citations
4,523
 
H-index
20
 
i10-index
25
 
Publications
20
 
Co-authors
15
list available
Contact
No contact links provided.
Publications
9 items
FreqEdit: Preserving High-Frequency Features for Robust Multi-Turn Image Editing
2025
Cited
0
GRPO-Guard: Mitigating Implicit Over-Optimization in Flow Matching via Regulated Clipping
2025
Cited
0
VR-Thinker: Boosting Video Reward Models through Thinking-with-Image Reasoning
2025
Cited
0
Scaling Image and Video Generation via Test-Time Evolutionary Search
2025
Cited
0
Flow-GRPO: Training Flow Matching Models via Online RL
2025
Cited
0
Asymmetric Decision-Making in Online Knowledge Distillation:Unifying Consensus and Divergence
2025
Cited
0
LEDiT: Your Length-Extrapolatable Diffusion Transformer without Positional Encoding
2025
Cited
0
Optimizing Knowledge Distillation in Transformers: Enabling Multi-Head Attention without Alignment Barriers
2025
Cited
0
Resume (English only)
Co-authors
15 total
Co-author 1
Co-author 1
Co-author 2
Co-author 2
Yao Tang
Yao Tang
Megvii
Haoqiang Fan
Haoqiang Fan
Megvii
Co-author 5
Co-author 5
Cong Yao
Cong Yao
Alibaba DAMO Academy
Shuchang Zhou
Shuchang Zhou
Megvii Inc.
Yuzhi Wang
Yuzhi Wang
Research Engineer @ Megvii Inc.