Search papers, labs, and topics across Lattice.
2
2
3
9
SpargeAttention2 achieves 95% attention sparsity in video diffusion models with a 16.2x speedup, proving that trainable sparse attention can significantly outperform training-free methods without sacrificing generation quality.
Achieve an 18.6x speedup in video diffusion models with 97% attention sparsity by learning how to route and combine sparse and linear attention, outperforming heuristic approaches.
