OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913030125649920 |
|---|---|
| author | Jiang, Haoyang Wang, Zekun Yi, Mingyang Li, Xiuyu Hu, Lanqing Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju |
| author_facet | Jiang, Haoyang Wang, Zekun Yi, Mingyang Li, Xiuyu Hu, Lanqing Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju |
| contents | The Diffusion Probabilistic Model (DPM) achieves remarkable performance in image generation, while its increasing parameter size and computational overhead hinder its deployment in practical applications. To improve this, the existing literature focuses on obtaining a smaller model with a fixed architecture through model compression. However, in practice, DPMs usually need to be deployed on various devices with different resource constraints, which leads to multiple compression processes, incurring significant overhead for repeated training. To obviate this, we propose a once-for-all (OFA) compression framework for DPMs that yields different subnetworks with various computations in a one-shot training manner. The existing OFA framework typically involves massive subnetworks with different parameter sizes, while such a huge candidate space slows the optimization. Thus, we propose to restrict the candidate subnetworks with a certain set of parameter sizes, where each size corresponds to a specific subnetwork. Specifically, to construct each subnetwork with a given size, we gradually allocate the maintained channels by their importance. Furthermore, we propose a reweighting strategy to balance the optimization process of different subnetworks. Experimental results show that our approach can produce compressed DPMs for various sizes with significantly lower training overhead while achieving satisfactory performance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_12668 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner Jiang, Haoyang Wang, Zekun Yi, Mingyang Li, Xiuyu Hu, Lanqing Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju Computer Vision and Pattern Recognition The Diffusion Probabilistic Model (DPM) achieves remarkable performance in image generation, while its increasing parameter size and computational overhead hinder its deployment in practical applications. To improve this, the existing literature focuses on obtaining a smaller model with a fixed architecture through model compression. However, in practice, DPMs usually need to be deployed on various devices with different resource constraints, which leads to multiple compression processes, incurring significant overhead for repeated training. To obviate this, we propose a once-for-all (OFA) compression framework for DPMs that yields different subnetworks with various computations in a one-shot training manner. The existing OFA framework typically involves massive subnetworks with different parameter sizes, while such a huge candidate space slows the optimization. Thus, we propose to restrict the candidate subnetworks with a certain set of parameter sizes, where each size corresponds to a specific subnetwork. Specifically, to construct each subnetwork with a given size, we gradually allocate the maintained channels by their importance. Furthermore, we propose a reweighting strategy to balance the optimization process of different subnetworks. Experimental results show that our approach can produce compressed DPMs for various sizes with significantly lower training overhead while achieving satisfactory performance. |
| title | OFA-Diffusion Compression: Compressing Diffusion Model in One-Shot Manner |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.12668 |