Neural Machine Translation of Rare Words with Subword Units

Neural machine translation (NMT) models typically operate with a fixed vocabulary, so the translation of rare and unknown words is an open problem. Previous work addresses this problem through back-off dictionaries. In this paper, we introduce a simpler and more effective approach, making the NMT model capable of open-vocabulary translation by encoding rare and unknown words as sequences of subword units, based on the intuition that various word classes are translatable via smaller units than words, for instance names (via character copying or transliteration), compounds (via compositional translation), and cognates and loanwords (via phonological and morphological transformations). We discuss the suitability of different word segmentation techniques, including simple character $n$-gram models and a segmentation based on the \emph{byte pair encoding} compression algorithm, and empirically show that subword models improve over a back-off dictionary baseline for the WMT 15 translation tasks English-German and English-Russian by 1.1 and 1.3 BLEU, respectively.