Fetching the paper…

CLIPErase: Efficient Unlearning of Visual-Textual Associations in CLIP · Around