Phrase-Based & Neural Unsupervised Machine Translation

EMNLP 2018 Guillaume LampleMyle OttAlexis ConneauLudovic DenoyerMarc'Aurelio Ranzato

Machine translation systems achieve near human-level performance on some languages, yet their effectiveness strongly relies on the availability of large amounts of parallel sentences, which hinders their applicability to the majority of language pairs. This work investigates how to learn to translate when having access to only large monolingual corpora in each language... (read more)

PDF Abstract

Evaluation results from the paper


Task Dataset Model Metric name Metric value Global rank Compare
Machine Translation WMT2014 English-French PBSMT + NMT BLEU score 27.60 # 25
Unsupervised Machine Translation WMT2014 English-French PBSMT + NMT BLEU 27.6 # 4
Machine Translation WMT2014 English-French Unsupervised NMT + Transformer BLEU score 25.14 # 28
Machine Translation WMT2014 English-French Unsupervised PBSMT BLEU score 28.11 # 24
Machine Translation WMT2014 English-German Unsupervised NMT + Transformer BLEU score 17.16 # 22
Machine Translation WMT2014 English-German PBSMT + NMT BLEU score 20.23 # 17
Machine Translation WMT2014 English-German Unsupervised PBSMT BLEU score 17.94 # 20
Unsupervised Machine Translation WMT2014 French-English PBSMT + NMT BLEU 27.7 # 4
Unsupervised Machine Translation WMT2016 English-German PBSMT + NMT BLEU 20.2 # 4
Machine Translation WMT2016 English-Romanian Unsupervised PBSMT BLEU score 21.33 # 9
Machine Translation WMT2016 English-Romanian Unsupervised NMT + Transformer BLEU score 21.18 # 10
Machine Translation WMT2016 English-Romanian PBSMT + NMT BLEU score 25.13 # 8
Machine Translation WMT2016 English-Russian PBSMT + NMT BLEU score 13.76 # 2
Machine Translation WMT2016 English-Russian Unsupervised PBSMT BLEU score 13.37 # 3
Machine Translation WMT2016 English-Russian Unsupervised NMT + Transformer BLEU score 7.98 # 4
Unsupervised Machine Translation WMT2016 German-English PBSMT BLEU 25.2 # 5