Denis Volkhonskiy

Skolkovo Institute of Science and Technology

Papers

Total Citations

H-Index

About

Denis Volkhonskiy is a researcher at the forefront of generative modeling, with a primary focus on video generation and efficient deep learning architectures. His most influential work, "Latent Video Transformer" (2021), tackles the critical challenge of high computational costs in video generation—where some models demand up to 512 TPUs for training—by proposing a more resource-efficient approach. This paper has garnered 14 citations, reflecting its relevance in a field hungry for scalable solutions. Volkhonskiy’s contributions lie in advancing latent-space transformers for video prediction, enabling the synthesis of future frames from past ones while reducing hardware barriers. His work bridges the gap between state-of-the-art generative performance and practical accessibility, making him a notable figure in the push toward democratized AI research. By addressing the computational bottleneck that limits many video models, Volkhonskiy has helped pave the way for more sustainable and widely applicable generative systems. His research continues to inspire efforts to balance model quality with efficiency, a crucial direction for the next generation of video AI.