Video Representation Learning with Joint-Embedding Predictive Architectures.

Katrina Drozdov, Ravid Shwartz-Ziv, Yann LeCun