Machine Translation on WMT2016 English-German

Metric: BLEU score (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

#	Model↕	BLEU score▼	Extra Data	Paper	Date↕	Code
1	MADL	40.68	No	-	-	-
2	Attentional encoder-decoder + BPE	34.2	No	Edinburgh Neural Machine Translation Systems for...	2016-06-09	Code
3	Linguistic Input Features	28.4	No	Linguistic Input Features Improve Neural Machine...	2016-06-09	Code
4	DeLighT	28	No	DeLighT: Deep and Light-weight Transformer	2020-08-03	Code
5	FLAN 137B (zero-shot)	27	No	Finetuned Language Models Are Zero-Shot Learners	2021-09-03	Code
6	Transformer	26.7	No	On the adequacy of untuned warmup for adaptive o...	2019-10-09	Code
7	FLAN 137B (few-shot, k=11)	26.1	No	Finetuned Language Models Are Zero-Shot Learners	2021-09-03	Code
8	BiRNN + GCN (Syn + Sem)	24.9	No	Exploiting Semantics in Neural Machine Translati...	2018-04-23	-
9	SMT + iterative backtranslation (unsupervised)	18.23	No	Unsupervised Statistical Machine Translation	2018-09-04	Code
10	Unsupervised NMT + weight-sharing	10.86	No	Unsupervised Neural Machine Translation with Wei...	2018-04-24	Code
11	Unsupervised S2S with attention	9.64	No	Unsupervised Machine Translation Using Monolingu...	2017-10-31	Code