LLM Unlearning via Loss Adjustment with Only Forget Data

Open in new window