Hardware
AdaCluster: Adaptive Query-Key Clustering for Sparse Attention in Video Generation
arXiv:2604.18348v1 Announce Type: new Abstract: Video diffusion transformers (DiTs) suffer from prohibitive inference latency due to quadratic attention complexity. Existing sparse attention methods e
arXiv:2604.18348v1 Announce Type: new Abstract: Video diffusion transformers (DiTs) suffer from prohibitive inference latency due to quadratic attention complexity. Existing sparse attention methods either overlook semantic similarity or fail to adapt to heterogeneous token distributions across layers, leading to model performance degradation. We propose AdaCluster, a training-free adaptive clustering framework that accelerates the generation of DiTs while preserving accuracy. AdaCluster applies an angle-similarity-preserving clustering method to query vectors for higher compression, and designs a euclidean-similarity-preserving clustering method for keys, covering cluster number assignment, threshold-wise adaptive clustering, and efficient critical cluster selection. Experiments on CogVideoX-2B, HunyuanVideo, and Wan-2.1 on one A40 GPU demonstrate up to 1.67-4.31x speedup with negligible quality degradation.
Related
- RainFusion2.0: Temporal-Spatial Awareness and Hardware-Efficient Block-wise Sparse Attention
- MaMe & MaRe: Matrix-Based Token Merging and Restoration for Efficient Visual Perception and Synthesis
- AdaSplash-2: Faster Differentiable Sparse Attention
- RACE Attention: A Strictly Linear-Time Attention Layer for Training on Outrageously Large Contexts
Source: arXiv cs.CV | 2026-04-21