On: Ask HN: What's the most creative 'useless' program you've ever written?

[IAmGraydon]

I’m pretty sure it is, except instead of comparing the result to a source image, it is analyzed and scored by a model that was trained to recognize images.

[Diti → IAmGraydon]

Diffusion models are trained to denoise.

[msephton → IAmGraydon]

It's kind of the opposite. Stable Diffusion starts with random noise, changes it, asks the model of it's representative of the goal, repeat. But there's much more to it.