CMLM+LAT+4 iterations

Reported on 4 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 Processing4 results

Machine TranslationonWMT2014 German-English
BLEU score· 2020-11-12
32.04
best: 35.15 (Bi-SimCut)
SOTA
Incorporating a Local Translation Mechanism into Non-autoregressive Translation arXiv:2011.06132
Machine TranslationonWMT2016 Romanian-English
BLEU score· 2020-11-12
33.26
best: 40.3 (fast-noisy-channel-modeling)
Incorporating a Local Translation Mechanism into Non-autoregressive Translation arXiv:2011.06132
Machine TranslationonWMT2014 English-German
BLEU score· 2020-11-12
27.35
best: 35.14 (Transformer Cycle (Rev))
Incorporating a Local Translation Mechanism into Non-autoregressive Translation arXiv:2011.06132
Machine TranslationonWMT2016 English-Romanian
BLEU score· 2020-11-12
32.87
best: 34.7 (DeLighT)
Incorporating a Local Translation Mechanism into Non-autoregressive Translation arXiv:2011.06132