Search papers, labs, and topics across Lattice.
1
0
3
8
Autoregressive video diffusion models can achieve faster decoding, lower memory footprint, and higher quality long-horizon generations by learning to attend to only the most salient spatiotemporal blocks.
