S^2-DMs:Skip-Step Diffusion Models
CoRR(2024)
摘要
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
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
resources is provided in the github.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要