Pavlo Molchanov
Scholar

Pavlo Molchanov

Google Scholar ID: J9PoyoIAAAAJ
NVIDIA Research
AIMachine LearningEfficient Deep LearningSemi-supervised learningnetwork inversion
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
14,116
 
H-index
45
 
i10-index
76
 
Publications
20
 
Co-authors
46
list available
Contact
No contact links provided.
Publications
39 items
Fast-dVLM: Efficient Block-Diffusion VLM via Direct Conversion from Autoregressive VLM
2026
Cited
0
Attend Before Attention: Efficient and Scalable Video Understanding via Autoregressive Gazing
2026
Cited
0
Stateful Token Reduction for Long-Video Hybrid VLMs
2026
Cited
0
Extending Puzzle for Mixture-of-Experts Reasoning Models with Application to GPT-OSS Acceleration
2026
Cited
0
C-RADIOv4 (Tech Report)
2026
Cited
0
GDPO: Group reward-Decoupled Normalization Policy Optimization for Multi-reward RL Optimization
arXiv.org · 2026
Cited
14
Efficient-DLM: From Autoregressive to Diffusion Language Models, and Beyond in Speed
2025
Cited
0
GSPN-2: Efficient Parallel Sequence Modeling
2025
Cited
0
Resume (English only)
Co-authors
46 total
Jan Kautz
Jan Kautz
Vice President of Research, NVIDIA Research
Hongxu (Danny) Yin
Hongxu (Danny) Yin
Research Scientist & Tech Lead, NVIDIA Research; Princeton University
Jose M. Alvarez
Jose M. Alvarez
NVIDIA
Umar Iqbal
Umar Iqbal
Research Manager at NVIDIA Research
Co-author 5
Co-author 5
Stephen Tyree
Stephen Tyree
NVIDIA
Greg Heinrich
Greg Heinrich
NVIDIA
Co-author 8
Co-author 8