Search papers, labs, and topics across Lattice.
3
0
4
3
A single distilled model can outperform larger heterogeneous teachers by effectively integrating their strengths without interference.
Bridging disparate model families, Any-OPD achieves a 4.5% increase in performance while reducing model size by 80%.
Unlock the full potential of your pretrained video diffusion models with a surprisingly simple four-stage post-training framework that drastically improves visual quality, temporal coherence, and instruction following.
