Learning Morphological Normalization for Translation from and into Morphologically Rich Languages

Burlot, Franck; Yvon, François

doi:10.1515/pralin-2017-0008

Cited by 4 publications

(5 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We used different costs for predicting lemmas and each tag, which are summed into a final objective function. As recently seen in (Martinez et al, 2016;Burlot and Yvon, 2017), these objectives are individually easier when working with morphologically-rich languages, and fully inflected words can be obtained by using morphological inflection models, which have been shown to be quite successful (Faruqui et al, 2016;Kann et al, 2017).…”

Section: Predicting Root and Tags Jointlymentioning

confidence: 94%

Character and Subword-Based Word Representation for Neural Language Modeling Prediction

Labeau¹,

Allauzen²

2017

Proceedings of the First Workshop on Subword and Character Level Models in NLP

View full text Add to dashboard Cite

Most of neural language models use different kinds of embeddings for word prediction. While word embeddings can be associated to each word in the vocabulary or derived from characters as well as factored morphological decomposition, these word representations are mainly used to parametrize the input, i.e. the context of prediction. This work investigates the effect of using subword units (character and factored morphological decomposition) to build output representations for neural language modeling. We present a case study on Czech, a morphologically-rich language, experimenting with different input and output representations. When working with the full training vocabulary, despite unstable training, our experiments show that augmenting the output word representations with character-based embeddings can significantly improve the performance of the model. Moreover, reducing the size of the output look-up table, to let the character-based embeddings represent rare words, brings further improvement.

show abstract

Section: Predicting Root and Tags Jointlymentioning

confidence: 94%

Character and Subword-Based Word Representation for Neural Language Modeling Prediction

Labeau¹,

Allauzen²

2017

Proceedings of the First Workshop on Subword and Character Level Models in NLP

View full text Add to dashboard Cite

show abstract

“…The specific setup we have used consisted in an architecture that enables training towards a dual objective: at each time-step in the output sentence, a normalized word and a PoStag are produced. To obtain the first factor vocabulary, all target words have been normalized (Burlot and Yvon, 2017a), i.e. all grammatical information that is redundant wrt.…”

Section: Kitmentioning

confidence: 99%

The QT21 Combined Machine Translation System for English to Latvian

Peter¹,

Ney²,

Bojar³

et al. 2017

Proceedings of the Second Conference on Machine Translation

Self Cite

View full text Add to dashboard Cite

This paper describes the joint submission of the QT21 projects for the English→Latvian translation task of the EMNLP 2017 Second Conference on Machine Translation (WMT 2017). The submission is a system combination which combines seven different statistical machine translation systems provided by the different groups.The systems are combined using either RWTH's system combination approach, or USFD's consensus-based systemselection approach. The final submission shows an improvement of 0.5 BLEU compared to the best single system on newstest2017.

show abstract

“…Normalization is usually performed using handcrafted rules and requires expert knowledge for each language pair. In this paper, normalized words are obtained with an automatic data-driven method 1 introduced in (Burlot and Yvon, 2017b).…”

Section: Normalizing Word Formsmentioning

confidence: 99%

Word Representations in Factored Neural Machine Translation

Burlot¹,

García-Martínez²,

Barrault³

et al. 2017

Proceedings of the Second Conference on Machine Translation

Self Cite

View full text Add to dashboard Cite

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.

show abstract

Learning Morphological Normalization for Translation from and into Morphologically Rich Languages

Cited by 4 publications

References 9 publications

Character and Subword-Based Word Representation for Neural Language Modeling Prediction

Character and Subword-Based Word Representation for Neural Language Modeling Prediction

The QT21 Combined Machine Translation System for English to Latvian

Word Representations in Factored Neural Machine Translation

Contact Info

Product

Resources

About