Abstract

The authors of (Cho et al., 2014a) have shown that the recently introduced neural network translation systems suffer from a significant drop in translation quality when translating long sentences, unlike existing phrase-based translation systems. In this paper, we propose a way to address this issue by automatically segmenting an input sentence into phrases that can be easily translated by the neural network translation model. Once each segment has been independently translated by the neural machine translation model, the translated clauses are concatenated to form a final translation. Empirical results show a significant improvement in translation quality for long sentences.

Highlights

  • Up to now, most research efforts in statistical machine translation (SMT) research have relied on the use of a phrase-based system as suggested in (Koehn et al, 2003)

  • One intuition is that the segmentation leads to multiple short clauses with less unknown words, which leads to more stable translation of each clause by the neural translation model

  • One interesting phenomenon is that any random segmentation was better than the direct translation without any segmentation. This indirectly agrees well with the previous finding in (Cho et al, 2014a) that the neural machine translation suffers from long sentences

Read more

Summary

Introduction

Most research efforts in statistical machine translation (SMT) research have relied on the use of a phrase-based system as suggested in (Koehn et al, 2003). An entirely new, neural network based approach has been proposed by several research groups (Kalchbrenner and Blunsom, 2013; Sutskever et al, 2014; Cho et al, 2014b), showing promising results, both as a standalone system or as an additional component in the existing phrase-based system. In this neural network based approach, an encoder ‘encodes’ a variable-length input sentence into a fixed-length vector and a decoder ‘decodes’ a variable-length target sentence from the fixedlength encoded vector. We show empirically that this approach improves translation quality of long sentences, compared to using a neural network to translate a whole sentence without segmentation

Background
Automatic Segmentation and Translation
Issues and Discussion
Dataset
Models and Approaches
Validity of the Automatic Segmentation
Importance of Using an Inverse Model
Quantitative and Qualitative Analysis
Discussion and Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call