Search papers, labs, and topics across Lattice.
University of Oregon
2
0
5
16
Ditch quadratic attention in your ViTs without sacrificing performance: ViT-AdaLA distills knowledge from pre-trained VFMs into linear attention architectures, achieving state-of-the-art results on classification and segmentation.
Finally, AI can generate hour-long videos with consistent characters and backgrounds, thanks to a new framework that nails seamless transitions between shots.
