Diffusion models have become the go-to method for many generative tasks, particularly for image-to-image generation tasks such as super-resolution and inpainting. Current diffusion-based methods do not provide statistical guarantees regarding the generated results, often preventing their use in high-stakes situations. To bridge this gap, we construct a confidence interval around each generated pixel such that the true value of the pixel is guaranteed to fall within the interval with a probability set by the user. Since diffusion models parametrize the data distribution, a straightforward way of constructing such intervals is by drawing multiple samples and calculating their bounds. However, this method has several drawbacks: i) slow sampling speeds ii) suboptimal bounds iii) requires training a diffusion model per task. To mitigate these shortcomings we propose Conffusion, wherein we fine-tune a pre-trained diffusion model to predict interval bounds in a single forward pass. We show that Conffusion outperforms the baseline method while being three orders of magnitude faster.
翻译:传播模型已成为许多基因化任务,特别是图像到图像生成任务,如超分辨率和油漆等的传导方法。当前基于扩散的方法不提供对所产生结果的统计保障,常常防止在高占用情况下使用这些结果。为了缩小这一差距,我们围绕每个生成的像素构建了一个信任间隔,保证像素的真实值在用户设定的概率范围内下降。由于扩散模型使数据分布平衡,构建这种间隔的一个直接方式是绘制多个样本并计算其界限。然而,这种方法有若干缺点:(一) 慢采样速度(二) 亚最佳界限(三) 需要每任务培训一个扩散模型。为了减轻这些缺点,我们提议合并,即我们微调一个经过预先训练的传播模型,以预测单个远端的间隔线。我们显示,聚合超越了基线方法,同时速度加快了三个数量级。