Machine Translation on WMT2016 Romanian-English

Metric: BLEU score (higher is better)

LeaderboardDataset

Loading chart...

Results

Hide extra data

Sort:

#	Model↕	BLEU score▼	Extra Data	Paper	Date↕	Code
1	fast-noisy-channel-modeling	40.3	Yes	Language Models not just for Pre-training: Fast ...	2020-11-13	Code
2	FLAN 137B (few-shot, k=9)	38.1	No	Finetuned Language Models Are Zero-Shot Learners	2021-09-03	Code
3	FLAN 137B (zero-shot)	37.3	No	Finetuned Language Models Are Zero-Shot Learners	2021-09-03	Code
4	MLM pretraining	35.3	No	Cross-lingual Language Model Pretraining	2019-01-22	Code
5	GenTranslate	33.5	No	GenTranslate: Large Language Models are Generati...	2024-02-10	Code
6	Attentional encoder-decoder + BPE	33.3	No	Edinburgh Neural Machine Translation Systems for...	2016-06-09	Code
7	Levenshtein Transformer (distillation)	33.26	No	Levenshtein Transformer	2019-05-27	Code
8	CMLM+LAT+4 iterations	33.26	No	Incorporating a Local Translation Mechanism into...	2020-11-12	Code
9	Adaptively Sparse Transformer (1.5-entmax)	33.1	No	Adaptively Sparse Transformers	2019-08-30	Code
10	HeadMask (Impt-18)	32.95	No	Alleviating the Inequality of Attention Heads fo...	2020-09-21	-
11	FlowSeq-large (NPD n = 30)	32.91	No	FlowSeq: Non-Autoregressive Conditional Sequence...	2019-09-05	Code
12	Adaptively Sparse Transformer (alpha-entmax)	32.89	No	Adaptively Sparse Transformers	2019-08-30	Code
13	HeadMask (Random-18)	32.85	No	Alleviating the Inequality of Attention Heads fo...	2020-09-21	-
14	FlowSeq-large (NPD n = 15)	32.46	No	FlowSeq: Non-Autoregressive Conditional Sequence...	2019-09-05	Code
15	FlowSeq-large (IWD n = 15)	32.03	No	FlowSeq: Non-Autoregressive Conditional Sequence...	2019-09-05	Code
16	NAT +FT + NPD	31.44	No	Non-Autoregressive Neural Machine Translation	2017-11-07	Code
17	CMLM+LAT+1 iterations	31.24	No	Incorporating a Local Translation Mechanism into...	2020-11-12	Code
18	FlowSeq-large	30.69	No	FlowSeq: Non-Autoregressive Conditional Sequence...	2019-09-05	Code
19	Denoising autoencoders (non-autoregressive)	30.3	No	Deterministic Non-Autoregressive Neural Sequence...	2018-02-19	Code
20	FlowSeq-base	30.16	No	FlowSeq: Non-Autoregressive Conditional Sequence...	2019-09-05	Code

#1fast-noisy-channel-modelingSOTA
40.3
BLEU score· Extra Data· 2020-11-13
Language Models not just for Pre-training: Fast Online Neural Noisy Channel Modeling Code
#2FLAN 137B (few-shot, k=9)
38.1
BLEU score· 2021-09-03
Finetuned Language Models Are Zero-Shot Learners Code
#3FLAN 137B (zero-shot)
37.3
BLEU score· 2021-09-03
Finetuned Language Models Are Zero-Shot Learners Code
#4MLM pretrainingSOTA
35.3
BLEU score· 2019-01-22
Cross-lingual Language Model Pretraining Code
#5GenTranslate
33.5
BLEU score· 2024-02-10
GenTranslate: Large Language Models are Generative Multilingual Speech and Machine Translators Code
#6Attentional encoder-decoder + BPESOTA
33.3
BLEU score· 2016-06-09
Edinburgh Neural Machine Translation Systems for WMT 16 Code
#7Levenshtein Transformer (distillation)
33.26
BLEU score· 2019-05-27
Levenshtein Transformer Code
#8CMLM+LAT+4 iterations
33.26
BLEU score· 2020-11-12
Incorporating a Local Translation Mechanism into Non-autoregressive Translation Code
#9Adaptively Sparse Transformer (1.5-entmax)
33.1
BLEU score· 2019-08-30
Adaptively Sparse Transformers Code
#10HeadMask (Impt-18)
32.95
BLEU score· 2020-09-21
Alleviating the Inequality of Attention Heads for Neural Machine Translation
#11FlowSeq-large (NPD n = 30)
32.91
BLEU score· 2019-09-05
FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow Code
#12Adaptively Sparse Transformer (alpha-entmax)
32.89
BLEU score· 2019-08-30
Adaptively Sparse Transformers Code
#13HeadMask (Random-18)
32.85
BLEU score· 2020-09-21
Alleviating the Inequality of Attention Heads for Neural Machine Translation
#14FlowSeq-large (NPD n = 15)
32.46
BLEU score· 2019-09-05
FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow Code
#15FlowSeq-large (IWD n = 15)
32.03
BLEU score· 2019-09-05
FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow Code
#16NAT +FT + NPD
31.44
BLEU score· 2017-11-07
Non-Autoregressive Neural Machine Translation Code
#17CMLM+LAT+1 iterations
31.24
BLEU score· 2020-11-12
Incorporating a Local Translation Mechanism into Non-autoregressive Translation Code
#18FlowSeq-large
30.69
BLEU score· 2019-09-05
FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow Code
#19Denoising autoencoders (non-autoregressive)
30.3
BLEU score· 2018-02-19
Deterministic Non-Autoregressive Neural Sequence Modeling by Iterative Refinement Code
#20FlowSeq-base
30.16
BLEU score· 2019-09-05
FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow Code