Kapil Krishnakumar
Papers
1
Total Citations
3
H-Index
1
About
Kapil Krishnakumar is a leading researcher in self-supervised learning and video understanding, with a focus on building AI systems that learn from observation rather than explicit labels. His most notable work, *V-JEPA 2* (2025), tackles a fundamental challenge in modern AI: enabling models to understand, predict, and plan in the physical world by leveraging internet-scale video data combined with minimal interaction data, such as robot trajectories. This approach advances the goal of developing agents that learn largely through passive observation, a paradigm shift toward more efficient and scalable AI. With 3 citations in its first year, *V-JEPA 2* is already influencing the intersection of computer vision and robotics. Krishnakumar’s contributions are pivotal for creating models that bridge the gap between perception and action, promising breakthroughs in autonomous systems and embodied AI. His work exemplifies how self-supervised techniques can unlock robust world models, making him a key figure to watch in the evolution of video-based learning and planning.
Research Focus
Key Achievements
Top Papers
- 1