Vector Space Models for Phrase-based Machine Translation

Abstract

This paper investigates the application of vector space models (VSMs) to the standard phrase-based machine translation pipeline. VSMs are models based on continuous word representations embedded in a vector space. We exploit word vectors to augment the phrase table with new inferred phrase pairs. This helps reduce out-of-vocabulary (OOV) words. In addition, we present a simple way to learn bilingually-constrained phrase vectors. The phrase vectors are then used to provide additional scoring of phrase pairs, which fits into the standard log-linear framework of phrase-based statistical machine translation. Both methods result in significant improvements over a competitive in-domain baseline applied to the Arabic-to-English task of IWSLT 2013.

Effective Cross-lingual Transfer of Neural Machine Translation Models without Shared Vocabularies

“…Cross-lingual word embedding is studied for the usages in MT as follows. In phrase-based SMT, Alkhouli et al (2014) builds translation models with word/phrase embeddings. Kim et al (2018) uses cross-lingual word embedding as a basic translation model for unsupervised MT and attach other components on top of it.…”

Using Word Embeddings for Improving Statistical Machine Translation of Phrasal Verbs

“…Following the work in Mikolov et al (2013b) and Alkhouli et al (2014), we introduce an additional feature in the translation model:…”

“…Following the work of Mikolov et al (2013a), Mikolov et al (2013b), and Alkhouli et al (2014), we exploit the idea that vector representations of similar words in different languages are related by a linear transformation. However, we focus on exploring this idea on a specific phenomenon with challenging semantics, namely PVs.…”

“…MWEs are either treated as a single unit or binary features encoding properties of MWEs are added to the translation table. On the other hand, recent works have successfully applied distributional representations of words and phrases in SMT (Mikolov et al (2013a), Zhang et al (2014), Alkhouli et al (2014)). The idea behind is that similar words and phrases in different languages tend to have simi-lar distributional representations (Mikolov et al, 2013a).…”

Improving Statistical Machine Translation with a Multilingual Paraphrase Database

“…Their model reaches comparable results to Saluja et al (2014) while works faster. Alkhouli et al (2014) use neural network phrase representation for paraphrasing OOVs and find translation for them using a phrase-table created from limited parallel data. Our experimental settings is different from the approaches in (Alkhouli et al, 2014;Mikolov et al, 2013a;Mikolov et al, 2013b).…”