zlacker

[parent] [thread] 0 comments
1. Taylor+(OP)[view] [source] 2022-05-24 00:07:53
My hunch is that they aren't tailored toward ridiculous images exactly, but if they demonstrated "a woman sitting in a chair reading", it would be really hard to tell if the result was a small modification of an image in the training data. If they demonstrate "A snake made out of corn", I have less concern about the model having a very close training example.
[go to top]