1
On the Hidden Costs of Counterfactual Knowledge Training in LLM Unlearning
反事实知识训练虽能帮助LLM遗忘特定信息，但论文揭示其背后的隐藏代价，值得关注。
arXiv:2605.27083v1 Announce Type: new Abstract: Counterfactual tuning (CFT) has emerged as a promising paradigm for Large Language Model (LLM) unlearn…