LLM Unlearning via Loss Adjustment with Only Forget Data

KELM
MU
Re-assign community