Machine Translation On Wmt2016 English German
평가 지표
BLEU score
평가 결과
이 벤치마크에서 각 모델의 성능 결과
비교 표
| 모델 이름 | BLEU score |
|---|---|
| finetuned-language-models-are-zero-shot | 26.1 |
| exploiting-monolingual-data-at-scale-for | - |
| finetuned-language-models-are-zero-shot | 27.0 |
| unsupervised-statistical-machine-translation | 18.23 |
| unsupervised-machine-translation-using | 9.64 |
| multi-agent-dual-learning | 40.68 |
| exploiting-semantics-in-neural-machine | 24.9 |
| on-the-adequacy-of-untuned-warmup-for | 26.7 |
| delight-very-deep-and-light-weight | 28.0 |
| edinburgh-neural-machine-translation-systems | 34.2 |
| unsupervised-neural-machine-translation-with-1 | 10.86 |
| linguistic-input-features-improve-neural | 28.4 |