Prompt Attacks Reveal Superficial Knowledge Removal in Unlearning Methods

AAMLMURe-assign community