News for hackers. Stuff that matters.
Re:Imagen, a text-to-image diffusion model by abel_ on Tuesday May 24, 2022 @06:01AM (#31490057)
On the contrary -- the opposite will happen. There's a decent body of research showing that just by training foundation models on their outputs, you amplify their capabilities.

Less common opinion: this is also how you end up with models that understand the concept of themselves, which has high economic value.

Even less common opinion: that's really dangerous.

There are no comments on this comment yet.