While conditional diffusion models are known to have good coverage of the
data distribution, they still face limitations in output diversity,
particularly when sampled with a high classifier-free guidance scale for
optimal image quality or when trained on small datasets. We attribute this
problem to the role of the conditioning signal in inference and offer an
improved sampling strategy for diffusion models that can increase generation
diversity, especially at high guidance scales, with minimal loss of sample
quality. Our sampling strategy anneals the conditioning signal by adding
scheduled, monotonically decreasing Gaussian noise to the conditioning vector
during inference to balance diversity and condition alignment. Our
Condition-Annealed Diffusion Sampler (CADS) can be used with any pretrained
model and sampling algorithm, and we show that it boosts the diversity of
diffusion models in various conditional generation tasks. Further, using an
existing pretrained diffusion model, CADS achieves a new state-of-the-art FID
of 1.70 and 2.31 for class-conditional ImageNet generation at 256$\times$256
and 512$\times$512 respectively.

通过使用一种改进的采样策略，我们提供了一种可以增加扩散模型的生成多样性的方法，尤其是在高指导尺度下，同时最小化样本质量损失。该方法通过在推理过程中向条件向量添加预定的、单调递减的高斯噪声来退火条件信号，以平衡多样性和条件对齐。我们的 Condition-Annealed Diffusion Sampler (CADS) 可以与任何预训练模型和采样算法一起使用，并且我们证明它在各种条件生成任务中提高了扩散模型的多样性。此外，通过使用现有的预训练扩散模型，CADS 在 256×256 和 512×512 像素的类条件 ImageNet 生成任务中取得了新的最优 FID 值，分别为 1.70 和 2.31。