News for hackers. Stuff that matters.
Re:“I've asked Stable Diffusion to generate 250 pages of 198... by orbital-decay on Thursday December 01, 2022 @07:57PM (#33825378)
Thing is, SD in particular is relatively small (860M parameters), and quantity can be somewhat transformed into quality in this case. As Google Parti landing page [0] conveniently demonstrates, more parameters with the same architecture yield more coherent output, including text and symbols. Given enough room (enough weights) in all parts of the model, starting from CLIP, it could even construct coherent text, as the compression ratio would be much less. However you'd need much beefier hardware to run it, and it might not be as efficient as having a better architecture or a dataset/training skewed towards symbols.

[0] https://parti.research.google/

There are no comments on this comment yet.