An Operation Sequence Model for Explainable Neural Machine Translation
This addresses the need for explainable machine translation in practical applications, though it is an incremental improvement over existing methods.
The paper tackles the problem of explainability in neural machine translation by proposing a model that generates target sentences through monotonic source traversal with explicit word alignment operations. The approach achieves competitive BLEU scores, outperforming a plain text system on Japanese-English and Portuguese-English and staying within 0.5 BLEU on Spanish-English.
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.