Today's denoising diffusion models do not "denoise" in the classical sense, i.e., they do not directly predict clean images. Rather, the neural networks predict noise or a noised quantity. In this paper, we suggest that predicting clean data and predicting noised quantities are fundamentally different. According to the manifold assumption, natural data should lie on a low-dimensional manifold, whereas noised quantities do not. With this assumption, we advocate for models that directly predict clean data, which allows apparently under-capacity networks to operate effectively in very high-dimensional spaces. We show that simple, large-patch Transformers on pixels can be strong generative models: using no tokenizer, no pre-training, and no extra loss. Our approach is conceptually nothing more than "Just image Transformers", or JiT, as we call it. We report competitive results using JiT with large patch sizes of 16 and 32 on ImageNet at resolutions of 256 and 512, where predicting high-dimensional noised quantities can fail catastrophically. With our networks mapping back to the basics of the manifold, our research goes back to basics and pursues a self-contained paradigm for Transformer-based diffusion on raw natural data.

翻译：当今的去噪扩散模型并非以经典意义上的“去噪”方式工作，即它们并不直接预测干净图像。相反，神经网络预测的是噪声或含噪量。本文认为，预测干净数据与预测含噪量存在本质区别。根据流形假设，自然数据应位于低维流形上，而含噪量则不然。基于此假设，我们倡导直接预测干净数据的模型，这使得表观容量不足的网络能够在极高维空间中有效运作。我们证明，在像素上使用简单的大块Transformer即可成为强大的生成模型：无需分词器、无需预训练、无需额外损失。我们的方法在概念上无非是“仅用图像Transformer”，我们称之为JiT。我们在ImageNet数据集上以256和512分辨率使用16和32的大块尺寸报告了JiT的竞争性结果，而在这些场景下预测高维含噪量可能导致灾难性失败。通过使网络回归流形的基本原理，我们的研究回归基础，追求一种基于Transformer的、在原始自然数据上进行扩散的自洽范式。