LSSF: Safety Alignment for Large Language Models through Low-Rank Safety Subspace Fusion.

Guanghao Zhou, Panjia Qiu, Cen Chen, Hongyu Li, Jason Chu, Xin Zhang, Jun Zhou

Browse the full ACL paper archive.