Machine Translation on WMT2014 English-German

Metric: SacreBLEU (higher is better)

LeaderboardDataset

Loading chart...

Results

Submit a result

Hide extra data

Sort:

#	Model↕	SacreBLEU▼	Extra Data	Paper	Date↕	Code
1	Noisy back-translation	33.8	Yes	Understanding Back-Translation at Scale	2018-08-28	Code
2	Transformer Cycle (Rev)	33.54	No	Lessons on Parameter Sharing across Layers in Tr...	2021-04-13	Code
3	Transformer+Rep(Uni)	32.35	No	Rethinking Perturbations in Encoder-Decoders for...	2021-04-05	Code
4	MAT	29.9	No	Multi-branch Attentive Transformer	2020-06-18	Code
5	Transformer (ADMIN init)	29.5	No	Very Deep Transformers for Neural Machine Transl...	2020-08-18	Code
6	Evolved Transformer Big	29.2	No	The Evolved Transformer	2019-01-30	Code
7	Mega	27.96	No	Mega: Moving Average Equipped Gated Attention	2022-09-21	Code

#1Noisy back-translationSOTA
33.8
SacreBLEU· Extra Data· 2018-08-28
Understanding Back-Translation at Scale Code
#2Transformer Cycle (Rev)
33.54
SacreBLEU· 2021-04-13
Lessons on Parameter Sharing across Layers in Transformers Code
#3Transformer+Rep(Uni)
32.35
SacreBLEU· 2021-04-05
Rethinking Perturbations in Encoder-Decoders for Fast Training Code
#4MAT
29.9
SacreBLEU· 2020-06-18
Multi-branch Attentive Transformer Code
#5Transformer (ADMIN init)
29.5
SacreBLEU· 2020-08-18
Very Deep Transformers for Neural Machine Translation Code
#6Evolved Transformer Big
29.2
SacreBLEU· 2019-01-30
The Evolved Transformer Code
#7Mega
27.96
SacreBLEU· 2022-09-21
Mega: Moving Average Equipped Gated Attention Code