One thing I find glossed over/lacking in the diffusion model materials is the contribution of the UNet.

Coming from someone that is just trying to catch up on what's going on, the UNet seems to play a huge role (if I understand right, this is where the convolutional neural networks are discovering 2d features.)

Relatively speaking, CNNs are kind of old news.. but they're a big deal. Unless I have something wrong? Do you know where I can learn more about how the UNet aspect works in depth?