Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2507.06739 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908442002718720 |
|---|---|
| author | Huang, Zishen Yang, Chunyu Ren, Mengyuan |
| author_facet | Huang, Zishen Yang, Chunyu Ren, Mengyuan |
| contents | Despite recent progress in video generation, inference speed remains a major bottleneck. A common acceleration strategy involves reusing model outputs via caching mechanisms at fixed intervals. However, we find that such fixed-frequency reuse significantly degrades quality in complex scenes, while manually tuning reuse thresholds is inefficient and lacks robustness. To address this, we propose Prompt-Complexity-Aware (PCA) caching, a method that automatically adjusts reuse thresholds based on scene complexity estimated directly from the input prompt. By incorporating prompt-derived semantic cues, PCA enables more adaptive and informed reuse decisions than conventional caching methods. We also revisit the assumptions behind TeaCache and identify a key limitation: it suffers from poor input-output relationship modeling due to an oversimplified prior. To overcome this, we decouple the noisy input, enhance the contribution of meaningful textual information, and improve the model's predictive accuracy through multivariate polynomial feature expansion. To further reduce computational cost, we replace the static CFGCache with DynCFGCache, a dynamic mechanism that selectively reuses classifier-free guidance (CFG) outputs based on estimated output variations. This allows for more flexible reuse without compromising output quality. Extensive experiments demonstrate that our approach achieves significant acceleration-for example, 2.79x speedup on the Wan2.1 model-while maintaining high visual fidelity across a range of scenes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_06739 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PromptTea: Let Prompts Tell TeaCache the Optimal Threshold Huang, Zishen Yang, Chunyu Ren, Mengyuan Computer Vision and Pattern Recognition Despite recent progress in video generation, inference speed remains a major bottleneck. A common acceleration strategy involves reusing model outputs via caching mechanisms at fixed intervals. However, we find that such fixed-frequency reuse significantly degrades quality in complex scenes, while manually tuning reuse thresholds is inefficient and lacks robustness. To address this, we propose Prompt-Complexity-Aware (PCA) caching, a method that automatically adjusts reuse thresholds based on scene complexity estimated directly from the input prompt. By incorporating prompt-derived semantic cues, PCA enables more adaptive and informed reuse decisions than conventional caching methods. We also revisit the assumptions behind TeaCache and identify a key limitation: it suffers from poor input-output relationship modeling due to an oversimplified prior. To overcome this, we decouple the noisy input, enhance the contribution of meaningful textual information, and improve the model's predictive accuracy through multivariate polynomial feature expansion. To further reduce computational cost, we replace the static CFGCache with DynCFGCache, a dynamic mechanism that selectively reuses classifier-free guidance (CFG) outputs based on estimated output variations. This allows for more flexible reuse without compromising output quality. Extensive experiments demonstrate that our approach achieves significant acceleration-for example, 2.79x speedup on the Wan2.1 model-while maintaining high visual fidelity across a range of scenes. |
| title | PromptTea: Let Prompts Tell TeaCache the Optimal Threshold |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.06739 |