Video World Models

Back to the Features: DINO as a Foundation for Video World Models
ICML Workshop 2025 - Learning physical world models in the latent space of DINOv2 from uncurated web videos.
Federico Baldassarre, Marc Szafraniec, Basile Terver, Vasil Khalidov, Francisco Massa, Yann LeCun, Patrick Labatut, Maximilian Seitzer, Piotr Bojanowski
PDF Cite
Back to the Features: DINO as a Foundation for Video World Models