论文标题
具有变压器的可扩展模型
Scalable Diffusion Models with Transformers
论文作者
论文摘要
我们根据变压器体系结构探索新的扩散模型。我们训练图像的潜在扩散模型，用在潜在斑块上运行的变压器代替常用的U-NET主链。我们通过GFLOPS测量的正向通行复杂性的镜头分析了扩散变压器（DIT）的可扩展性。我们发现，通过增加变压器的深度/宽度或输入令牌数量增加的DIT始终具有较低的FID。除了具有良好的可伸缩性外，我们最大的DIT-XL/2模型在类条件成像网512x512和256x256基准上优于所有先前的扩散模型，实现后者的最先进的FID为2.27。
We explore a new class of diffusion models based on the transformer architecture. We train latent diffusion models of images, replacing the commonly-used U-Net backbone with a transformer that operates on latent patches. We analyze the scalability of our Diffusion Transformers (DiTs) through the lens of forward pass complexity as measured by Gflops. We find that DiTs with higher Gflops -- through increased transformer depth/width or increased number of input tokens -- consistently have lower FID. In addition to possessing good scalability properties, our largest DiT-XL/2 models outperform all prior diffusion models on the class-conditional ImageNet 512x512 and 256x256 benchmarks, achieving a state-of-the-art FID of 2.27 on the latter.
