Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2505.21114 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912397877313536 |
|---|---|
| author | Wang, Shuai Li, Zexian zhang, Qipeng Song, Tianhui Li, Xubin Ge, Tiezheng Zheng, Bo Wang, Limin |
| author_facet | Wang, Shuai Li, Zexian zhang, Qipeng Song, Tianhui Li, Xubin Ge, Tiezheng Zheng, Bo Wang, Limin |
| contents | Diffusion models have demonstrated remarkable generation quality but at the cost of numerous function evaluations. Recently, advanced ODE-based solvers have been developed to mitigate the substantial computational demands of reverse-diffusion solving under limited sampling steps. However, these solvers, heavily inspired by Adams-like multistep methods, rely solely on t-related Lagrange interpolation. We show that t-related Lagrange interpolation is suboptimal for diffusion model and reveal a compact search space comprised of time steps and solver coefficients. Building on our analysis, we propose a novel differentiable solver search algorithm to identify more optimal solver. Equipped with the searched solver, rectified-flow models, e.g., SiT-XL/2 and FlowDCN-XL/2, achieve FID scores of 2.40 and 2.35, respectively, on ImageNet256 with only 10 steps. Meanwhile, DDPM model, DiT-XL/2, reaches a FID score of 2.33 with only 10 steps. Notably, our searched solver outperforms traditional solvers by a significant margin. Moreover, our searched solver demonstrates generality across various model architectures, resolutions, and model sizes. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_21114 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Differentiable Solver Search for Fast Diffusion Sampling Wang, Shuai Li, Zexian zhang, Qipeng Song, Tianhui Li, Xubin Ge, Tiezheng Zheng, Bo Wang, Limin Computer Vision and Pattern Recognition Diffusion models have demonstrated remarkable generation quality but at the cost of numerous function evaluations. Recently, advanced ODE-based solvers have been developed to mitigate the substantial computational demands of reverse-diffusion solving under limited sampling steps. However, these solvers, heavily inspired by Adams-like multistep methods, rely solely on t-related Lagrange interpolation. We show that t-related Lagrange interpolation is suboptimal for diffusion model and reveal a compact search space comprised of time steps and solver coefficients. Building on our analysis, we propose a novel differentiable solver search algorithm to identify more optimal solver. Equipped with the searched solver, rectified-flow models, e.g., SiT-XL/2 and FlowDCN-XL/2, achieve FID scores of 2.40 and 2.35, respectively, on ImageNet256 with only 10 steps. Meanwhile, DDPM model, DiT-XL/2, reaches a FID score of 2.33 with only 10 steps. Notably, our searched solver outperforms traditional solvers by a significant margin. Moreover, our searched solver demonstrates generality across various model architectures, resolutions, and model sizes. |
| title | Differentiable Solver Search for Fast Diffusion Sampling |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2505.21114 |