https://yanaiela.github.io/
arxiv.org/abs/2407.00211
arxiv.org/abs/2407.00211
We had a (somewhat) similar investigation in text2image models where we actually found that newer models are becoming even less diverse
arxiv.org/abs/2410.22592
We had a (somewhat) similar investigation in text2image models where we actually found that newer models are becoming even less diverse
arxiv.org/abs/2410.22592