Actually this is wrong, after reading the paper: https://arxiv.org/pdf/2302.03027.pdf https://arxiv.org/pdf/2302.03027.pdf The main idea is to specify an edit

Actually this is wrong, after reading the paper: https://arxiv.org/pdf/2302.03027.pdf https://arxiv.org/pdf/2302.03027.pdf The main idea is to specify an edit direction which is a broad statement of what you want to do. The model then uses gpt-3 to generate text for that direction and then it calculates embeddings with CLIP and then the mean difference. The difference is used to guide the noise to make the final change. One pitfall seems like you can't specify very customized changes, but very cool approach