Adaptively Sparse Transformer (1.5-entmax)

Reported on 3 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 Processing3 results

Machine TranslationonWMT2016 Romanian-English
BLEU score· 2019-08-30
33.1
best: 40.3 (fast-noisy-channel-modeling)
Adaptively Sparse Transformers arXiv:1909.00015
Machine TranslationonIWSLT2017 German-English
BLEU score· 2019-08-30
29.83
best: 29.9 (Adaptively Sparse Transformer (alpha-entmax))
Adaptively Sparse Transformers arXiv:1909.00015
Machine TranslationonWMT2014 English-German
BLEU score· 2019-08-30
25.89
best: 35.14 (Transformer Cycle (Rev))
Adaptively Sparse Transformers arXiv:1909.00015