An Operation Sequence Model for Explainable Neural Machine Translation

Conference Object
Change log
Byrne, Bill 

We propose to achieve explainable neural machine translation (NMT) by changing the output representation to explain itself. We present a novel approach to NMT which generates the target sentence by monotonically walking through the source sentence. Word reordering is modeled by operations which allow setting markers in the target sentence and move a target-side write head between those markers. In contrast to many modern neural models, our system emits explicit word alignment information which is often crucial to practical machine translation as it improves explainability. Our technique can outperform a plain text system in terms of BLEU score under the recent Transformer architecture on Japanese-English and Portuguese-English, and is within 0.5 BLEU difference on Spanish-English.

cs.CL, cs.CL
Journal Title
EMNLP BlackboxNLP workshop 2018
Conference Name
EMNLP BlackboxNLP workshop 2018
Journal ISSN
Volume Title
All rights reserved
EPSRC (1632937)
Engineering and Physical Sciences Research Council (EP/L027623/1)