Self-supervised and Supervised Joint Training for
Resource-rich Machine Translation
Yong Cheng 1 Wei Wang Lu Jiang 1 2 Wolfgang Macherey 1
Abstract supervised task on abundant unlabeled data (i.e. monolin-
gual sentences). In the second stage, the full or partial model
Self-supervised pre-training of text represen-
is fine ...


雷达卡




京公网安备 11010802022788号







