Internal Value Alignment in Large Language Models through Controlled Value Vector Activation.

Haoran Jin, Meng Li, Xiting Wang, Zhihao Xu, Minlie Huang, Yantao Jia, Defu Lian

Browse the full ACL paper archive.