Sol-Attn: Accelerating Video Generation Inference via On-the-Fly Attention Sparsification
- Published
- Source
- arXiv
- Paper number
- 731
- Field
- Computer Vision
- arXiv ID
- 2607.24027
Key points
- It speeds up a pretrained model without training by sparsifying only the attention.
- Block proxy scores are modeled as Gaussian, so blocks are selected using thresholds based on the mean and standard deviation.
- It does not build a proxy-score map in memory, which reduces routing cost.
- It reuses discarded block scores for approximate correction, so accuracy loss stays small even under aggressive sparsification.
- It achieves end-to-end speedups of 2.1x for video generation and 2.3x for editing, without degrading quality.
- When integrated with Sol-Engine, it reaches up to 5x acceleration for video generation.
Paper links
External research summaries. These are not HDATF publications or measured product results.