TY  - RPRT
TI  - A safety realignment framework via subspace-oriented model fusion for large language models
AU  - Xin Yi
AU  - Shunfan Zheng
AU  - Linlin Wang
AU  - Xiaoling Wang
AU  - Liang He
PY  - 2024
UR  - https://arxiv.org/abs/2405.09055
ID  - 2405.09055
ER  -