An Empirical Analysis of NMT-Derived Interlingual Embeddings and Their Use in Parallel Sentence Identification
Cristina España-Bonet, Ádám Varga, Alberto Barrón‐Cedeño, Josef van Genabith · IEEE Journal of Selected Topics in Signal Processing · 2017
End-to-end neural machine translation has overtaken statistical machine translation in terms of translation quality for some language pairs, specially those with large amounts of parallel data. Besides this palpable improvement, neural networks provide several new properties. A single system can be trained to translate between many languages at almost no additional cost other than training time. Furthermore, internal representations learned by the network serve as a new semantic representation of words-or sentences-which, unlike standard word embeddings, are learned in an essentially bilingual or even multilingual context. In view of these properties, the contribution of the present paper is twofold. First, we systematically study the neural machine translation (NMT) context vectors, i.e., output of the encoder, and their power as an interlingua representation of a sentence. We assess their quality and effectiveness by measuring similarities across translations, as well as semantically related and semantically unrelated sentence pairs. Second, as extrinsic evaluation of the first point, we identify parallel sentences in comparable corpora, obtaining an F1=98.2% on data from a shared task when using only NMT context vectors. Using context vectors jointly with similarity measures F1reaches 98.9%.