Syntactically Guided Neural Machine Translation

TL;DR

This paper introduces a method that combines hierarchical phrase-based SMT lattices with neural machine translation, improving decoding efficiency and translation quality, especially for large vocabularies.

Contribution

It proposes a novel approach to integrate Hiero scores into NMT decoding using weight pushing, enhancing translation performance and scalability.

Findings

Gains over Hiero and NMT decoding alone.

Effective extension of NMT to large vocabularies.

Practical improvements in translation quality.

Abstract

We investigate the use of hierarchical phrase-based SMT lattices in end-to-end neural machine translation (NMT). Weight pushing transforms the Hiero scores for complete translation hypotheses, with the full translation grammar score and full n-gram language model score, into posteriors compatible with NMT predictive probabilities. With a slightly modified NMT beam-search decoder we find gains over both Hiero and NMT decoding alone, with practical advantages in extending NMT to very large input and output vocabularies.