paper-with-me

홈 › Papers

S$^{2}$-DMs:Skip-Step Diffusion Models

2024-01-03 · YiXuan Wang, Shuangyin Li

Diffusion models have emerged as powerful generative tools, rivaling GANs in sample quality and mirroring the likelihood scores of autoregressive models. A subset of these models, exemplified by DDIMs, exhibit an inherent asymmetry: they are trained over $T$ steps but only sample from a subset of $T$ during generation. This selective sampling approach, though optimized for speed, inadvertently misses out on vital information from the unsampled steps, leading to potential compromises in sample quality. To address this issue, we present the S$^{2}$-DMs, which is a new training method by using an innovative $L_{skip}$, meticulously designed to reintegrate the information omitted during the selective sampling phase. The benefits of this approach are manifold: it notably enhances sample quality, is exceptionally simple to implement, requires minimal code modifications, and is flexible enough to be compatible with various sampling algorithms. On the CIFAR10 dataset, models trained using our algorithm showed an improvement of 3.27% to 14.06% over models trained with traditional methods across various sampling algorithms (DDIMs, PNDMs, DEIS) and different numbers of sampling steps (10, 20, ..., 1000). On the CELEBA dataset, the improvement ranged from 8.97% to 27.08%. Access to the code and additional resources is provided in the github.

📄 PDF Abstract BibTeX arXiv:2401.01520

Code (1)

kingkingofall/skip-step-diffusion 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Pretrained Diffusion Models Are Inherently Skipped-Step Samplers

2025-08-21 · Wenju Xu arxiv

Diffusion models have been achieving state-of-the-art results across various generation tasks. However, a notable drawback is their sequential generation process, requiring long-sequence step-by-step generation. Existing…

Memory-Efficient Fine-Tuning Diffusion Transformers via Dynamic Patch Sampling and Block Skipping

2026-03-21 · Sunghyun Park, Jeongho Kim, Hyoungwoo Park, Debasmit Das 외 arxiv

Diffusion Transformers (DiTs) have significantly enhanced text-to-image (T2I) generation quality, enabling high-quality personalized content creation. However, fine-tuning these models requires substantial computational …

LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers

2024-12-17 · Xuan Shen, Zhao Song, Yufa Zhou, Bo Chen 외

Diffusion Transformers have emerged as the preeminent models for a wide array of generative tasks, demonstrating superior performance and efficacy across various applications. The promising results come at the cost of sl…

Denoising

Training-Free Adaptive Diffusion with Bounded Difference Approximation Strategy

2024-10-13 · Hancheng Ye, Jiakang Yuan, Renqiu Xia, Xiangchao Yan 외

Diffusion models have recently achieved great success in the synthesis of high-quality images and videos. However, the existing denoising techniques in diffusion models are commonly based on step-by-step noise prediction…

DenoisingPrediction

The Surprising Effectiveness of Skip-Tuning in Diffusion Sampling

2024-02-23 · Jiajun Ma, Shuchen Xue, Tianyang Hu, Wenjia Wang 외

With the incorporation of the UNet architecture, diffusion probabilistic models have become a dominant force in image generation tasks. One key design in UNet is the skip connections between the encoder and decoder block…

DecoderImage Generation