Transformer (ADMIN init)

Reported on 4 benchmarks across 1 task · 1 paper

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 English-German
BLEU score· 2020-08-18
30.1
best: 35.14 (Transformer Cycle (Rev))
Very Deep Transformers for Neural Machine Translation arXiv:2008.07772
Machine TranslationonWMT2014 English-German
SacreBLEU· 2020-08-18
29.5
best: 33.8 (Noisy back-translation)
Very Deep Transformers for Neural Machine Translation arXiv:2008.07772
Machine TranslationonWMT2014 English-French
BLEU score· 2020-08-18
43.8
best: 46.4 (Transformer+BT (ADMIN init))
Very Deep Transformers for Neural Machine Translation arXiv:2008.07772
Machine TranslationonWMT2014 English-French
SacreBLEU· 2020-08-18
41.8
best: 44.4 (Transformer+BT (ADMIN init))
Very Deep Transformers for Neural Machine Translation arXiv:2008.07772