TY  - RPRT
TI  - Adding Interpretable Attention to Neural Translation Models Improves Word Alignment
AU  - Thomas Zenkel
AU  - Joern Wuebker
AU  - John DeNero
PY  - 2019
UR  - https://arxiv.org/abs/1901.11359
ID  - 1901.11359
ER  -