作者位次 #6/10ACL 2026Poster· Natural Language Processing
CAP : Controllable Alignment Prompting for Unlearning in LLM s
Zhaokun Wang · Jinyu Guo · Jingwen Pu · Hongli Pu · Meng Yang · Xunlei Chen · Jie Ou · Wenyi Li · Guangchun Luo · Wenhong Tian
摘要
Large language models (LLMs) trained on unfiltered corpora inherently risk retaining sensitive information, necessitating selective knowledge unlearning for regulatory compliance and ethical safety. However, existing parameter-modifying methods face fundamental limitations: high computational cos...