Command Palette
Search for a command to run...
Unsupervised Machine Translation On Wmt2014 1
評価指標
BLEU
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
| Paper Title | ||
|---|---|---|
| GPT-3 175B (Few-Shot) | 39.2 | Language Models are Few-Shot Learners |
| MASS (6-layer Transformer) | 34.9 | MASS: Masked Sequence to Sequence Pre-training for Language Generation |
| SMT + NMT (tuning and joint refinement) | 33.5 | An Effective Approach to Unsupervised Machine Translation |
| MLM pretraining for encoder and decoder | 33.3 | Cross-lingual Language Model Pretraining |
| SMT as posterior regularization | 28.9 | Unsupervised Neural Machine Translation with SMT as Posterior Regularization |
| PBSMT + NMT | 27.7 | Phrase-Based & Neural Unsupervised Machine Translation |
| SMT | 25.9 | Unsupervised Statistical Machine Translation |
0 of 7 row(s) selected.