TY  - RPRT
TI  - Leveraging Importance Sampling to Detach Alignment Modules from Large Language Models
AU  - Yi Liu
AU  - Dianqing Liu
AU  - Mingye Zhu
AU  - Junbo Guo
AU  - Yongdong Zhang
AU  - Zhendong Mao
PY  - 2025
UR  - https://arxiv.org/abs/2505.19700
ID  - 2505.19700
ER  -