Search papers, labs, and topics across Lattice.
2
0
4
2
Cosmos 3 sets a new benchmark for omnimodal models, outperforming existing state-of-the-art in both Text-to-Image and Image-to-Video tasks.
Ditch the action tokens: representing robot actions as interpretable, pixel-grounded "action images" lets your video backbone act as a zero-shot policy.
