← Search

Dengji Guo

2 accepted papers

2022

Prediction Difference Regularization against Perturbation for Neural Machine Translation

ACL 2022long

Regularization methods applying input perturbation have drawn considerable attention and have been frequently explored for NMT tasks in recent years. Despite their simplicity and effectiveness, we argue that these methods are limited by the under-fitting of training data. In this paper, we utilize p…

Cited by 15SourcePDFScholar
2021

Guiding Teacher Forcing with Seer Forcing for Neural Machine Translation

ACL 2021long

Although teacher forcing has become the main training paradigm for neural machine translation, it usually makes predictions only conditioned on past information, and hence lacks global planning for the future. To address this problem, we introduce another decoder, called seer decoder, into the encod…