Word Representations in Factored Neural Machine Translation

Translation into a morphologically rich language requires a large output vocabulary to model various morphological phenomena, which is a challenge for neural machine translation architectures. To address this issue, the present paper investigates the impact of having two output factors with a system able to generate separately two distinct representations of the target words. Within this framework, we investigate several word representations that correspond to different distributions of morpho-syntactic information across both factors. We report experiments for translation from English into two morphologically rich languages, Czech and Latvian, and show the importance of explicitly modeling target morphology.

• Publication year

  2017

• Venue

  Conference on Machine Translation

• Publication date

  2017-09-01

• Fields of study

  Linguistics, Computer Science

• Identifiers
  DOI 10.18653/v1/W17-4703
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Learning Morphological Normalization for Translation from and into Morphologically Rich Languages

  2017cited by this paper
• Evaluating the morphological competence of Machine Translation Systems

  2017cited by this paper
• NMTPY: A Flexible Toolkit for Advanced Neural Machine Translation Systems

  2017cited by this paper
• Stronger Baselines for Trustable Results in Neural Machine Translation

  2017cited by this paper
• Character-based Neural Machine Translation

  2016cited by this paper
• Guided Alignment Training for Topic-Aware Neural Machine Translation

  2016cited by this paper
• Linguistic Input Features Improve Neural Machine Translation

  2016cited by this paper
• Achieving Open Vocabulary Neural Machine Translation with Hybrid Word-Character Models

  2016cited by this paper
• Two-Step MT: Predicting Target Morphology

  2016cited by this paper
• Factored Neural Machine Translation Architectures

  2016cited by this paper
• CharacTer: Translation Edit Rate on Character Level

  2016cited by this paper
• Multi-Way, Multilingual Neural Machine Translation with a Shared Attention Mechanism

  2016cited by this paper
• Using Factored Word Representation in Neural Network Language Models

  2016cited by this paper
• Results of the WMT16 Metrics Shared Task

  2016cited by this paper
• Neural Machine Translation of Rare Words with Subword Units

  2015influential reference
• Multi-Task Learning for Multiple Language Translation

  2015cited by this paper
• Improving Neural Machine Translation Models with Monolingual Data

  2015influential reference
• Multi-task Sequence to Sequence Learning

  2015cited by this paper
• On Using Very Large Target Vocabulary for Neural Machine Translation

  2014cited by this paper
• Fitting Sentence Level Translation Evaluation with Many Dense Features

  2014cited by this paper
• Adam: A Method for Stochastic Optimization

  2014cited by this paper
• Neural Machine Translation by Jointly Learning to Align and Translate

  2014cited by this paper
• On the Properties of Neural Machine Translation: Encoder–Decoder Approaches

  2014cited by this paper
• Open-Source Tools for Morphology, Lemmatization, POS Tagging and Named Entity Recognition

  2014influential reference
• Using subcategorization knowledge to improve case prediction for translation to German

  2013cited by this paper
• Morphological Analysis with Limited Resources: Latvian Example

  2013cited by this paper
• XenC: An Open-Source Tool for Data Selection in Natural Language Processing

  2013cited by this paper
• Factored recurrent neural network language model in TED lecture transcription

  2012cited by this paper
• Probes in a Taxonomy of Factored Phrase-Based Models

  2012cited by this paper
• Translate, Predict or Generate: Modeling Rich Morphology in Statistical Machine Translation

  2012cited by this paper
• Rich Morphology Generation Using Statistical Machine Translation

  2012cited by this paper
• Modeling Inflection and Word-Formation in SMT

  2012cited by this paper
• 2010 Failures in English-Czech Phrase-Based MT

  2010cited by this paper
• Understanding the difficulty of training deep feedforward neural networks

  2010cited by this paper
• Applying Morphology Generation Models to Machine Translation

  2008cited by this paper
• Limsi’s Statistical Translation Systems for WMT‘08

  2008cited by this paper
• A Scalable Hierarchical Distributed Language Model

  2008cited by this paper
• Generating Complex Morphology for Machine Translation

  2007cited by this paper
• English-to-Czech Factored Machine Translation

  2007cited by this paper
• Factored Neural Language Models

  2006cited by this paper
• Bleu: a Method for Automatic Evaluation of Machine Translation

  2002influential reference

• Low-resource Multilingual Neural Translation Using Linguistic Feature-based Relevance Mechanisms

  2023cites this paper
• On the Use of Morpho-Syntactic Description Tags in Neural Machine Translation with Small and Large Training Corpora

  2022cites this paper
• Modeling Target-Side Morphology in Neural Machine Translation: A Comparison of Strategies

  2022influential citation
• Survey of Low-Resource Machine Translation

  2021cites this paper
• Sparsely Factored Neural Machine Translation

  2021cites this paper
• Arabic Machine Translation: A survey of the latest trends and challenges

  2020cites this paper
• Neural Machine Translation

  2020cites this paper
• Addressing data sparsity for neural machine translation between morphologically rich languages

  2020cites this paper
• Improving Low-Resource NMT through Relevance Based Linguistic Features Incorporation

  2020cites this paper
• The Translation Problem

  2020cites this paper
• Uses of Machine Translation

  2020cites this paper
• Neural Translation Models

  2020cites this paper
• Beyond Parallel Corpora

  2020cites this paper
• Low-Resource Machine Translation using Interlinear Glosses

  2019cites this paper
• POS Tag-enhanced Coarse-to-fine Attention for Neural Machine Translation

  2019cites this paper
• Using Interlinear Glosses as Pivot in Low-Resource Multilingual Machine Translation.

  2019cites this paper
• Neural Morphological Tagging of Lemma Sequences for Machine Translation

  2018cites this paper
• Neural Language Models

  2018cites this paper
• Tailoring Neural Architectures for Translating from Morphologically Rich Languages

  2018cites this paper
• LIMSI@WMT’17

  2017cites this paper
• The QT21 Combined Machine Translation System for English to Latvian

  2017cites this paper