Neural Machine Translation on IWSLT14 English-to-French (tst2012)
42.64BLEUCE+BT
Evaluation results
| Method | Links | |
|---|---|---|
| CE+BTLoss function=Cross-Entropy, Data augmentation=BT2021.05 | 42.64 | |
| WD+OURSLoss function=L2 Wasserstein distance, Data augmentation=OURS2021.05 | 41.07 | |
| CE+OURSLoss function=Cross-Entropy, Data augmentation=OURS2021.05 | 40.94 | |
| CELoss function=Cross-Entropy, Data augmentation=None2021.05 | 40.86 | |
| CE+MLMLoss function=Cross-Entropy, Data augmentation=MLM2021.05 | 40.79 | |
| WD+MLMLoss function=L2 Wasserstein distance, Data augmentation=MLM2021.05 | 40.61 | |
| WD+BTLoss function=L2 Wasserstein distance, Data augmentation=BT2021.05 | 40.6 | |
| WDLoss function=L2 Wasserstein distance, Data augmentation=None2021.05 | 40.48 |