Learn to Refuse: Making Large Language Models More Controllable and Reliable through Knowledge Scope Limitation and Refusal Mechanism.

Lang Cao

Browse the full EMNLP paper archive.