Unsupervised Machine Translation On Wmt2014 1
Métriques
BLEU
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | BLEU |
|---|---|
| language-models-are-few-shot-learners | 39.2 |
| mass-masked-sequence-to-sequence-pre-training | 34.9 |
| an-effective-approach-to-unsupervised-machine | 33.5 |
| unsupervised-statistical-machine-translation | 25.9 |
| unsupervised-neural-machine-translation-with | 28.9 |
| cross-lingual-language-model-pretraining | 33.3 |
| phrase-based-neural-unsupervised-machine | 27.7 |