NAT +FT + NPD

Reported on 5 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Machine TranslationonWMT2014 German-English
BLEU score· 2017-11-07
23.2
best: 35.15 (Bi-SimCut)
SOTA
Non-Autoregressive Neural Machine Translation arXiv:1711.02281
Machine TranslationonIWSLT2015 English-German
BLEU score· 2017-11-07
28.16
best: 30 (PS-KD)
Non-Autoregressive Neural Machine Translation arXiv:1711.02281
Machine TranslationonWMT2016 Romanian-English
BLEU score· 2017-11-07
31.44
best: 40.3 (fast-noisy-channel-modeling)
Non-Autoregressive Neural Machine Translation arXiv:1711.02281
Machine TranslationonWMT2014 English-German
BLEU score· 2017-11-07
19.17
best: 35.14 (Transformer Cycle (Rev))
Non-Autoregressive Neural Machine Translation arXiv:1711.02281
Machine TranslationonWMT2016 English-Romanian
BLEU score· 2017-11-07
29.79
best: 34.7 (DeLighT)
Non-Autoregressive Neural Machine Translation arXiv:1711.02281