Concept-ROT: Poisoning Concepts in Large Language Models with Model Editing.

Keltin Grimes, Marco Christiani, David Shriver, Marissa Catherine Connor

Browse the full ICLR paper archive.