两个并行采样器提升有限步数下扩散模型的生成质量。
It Takes Two to Tango: Two Parallel Samplers Improve Quality in Diffusion Models for Limited Steps
- 用两个并行采样器在连续时间点进行去噪，信息融合到隐空间。
- 在有限步数下，样本质量显著优于单采样器或简单拼接方案。
- 方法即插即用、无需微调，适用于各类扩散模型。
当扩散模型可用的去噪步骤有限时，我们发现两个并行采样器能提升生成图像的质量。两个采样器在连续时间点执行去噪操作，其信息被合理地整合进隐变量空间。该方法在概念和实现上均极为简洁：可即插即用、与模型无关，无需额外微调或外部模型。我们在多种扩散模型上通过自动评估和人工评测验证了该方法的有效性。同时发现，简单地将两个采样器的信息拼接会降低样本质量。此外，增加更多并行采样器并不必然带来质量提升。
原文摘要 · Abstract (English)
We consider the situation where we have a limited number of denoising steps, i.e., of evaluations of a diffusion model. We show that two parallel processors or samplers under such limitation can improve the quality of the sampled image. Particularly, the two samplers make denoising steps at successive times, and their information is appropriately integrated in the latent image. Remarkably, our method is simple both conceptually and to implement: it is plug-&-play, model agnostic, and does not require any additional fine-tuning or external models. We test our method with both automated and human evaluations for different diffusion models. We also show that a naive integration of the information from the two samplers lowers sample quality. Finally, we find that adding more parallel samplers does not necessarily improve sample quality.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。