Title《Pay Less Attention With Lightweight and Dynamic Convolutions](https://arxiv.org/abs/1901.10430)》 Abstract self-att

Title《MUSE: PARALLEL MULTI-SCALE ATTENTION FOR SEQUENCE TO SEQUENCE LEARNING》 AbstractQ:attention mechanism alone suffer

Title:《Very Deep Transformers for Neural Machine Translation》 Q：how to decrease the variance of the output layer in orde

title:《Understanding Back-Translation at Scale》Q:how to improve neural machine translation with monolingual data? S:augm