NICT_LOGO.JPG KYOTO-U_LOGO.JPG

WAT

The Workshop on Asian Translation
Evaluation Results

[EVALUATION RESULTS TOP] | [BLEU] | [RIBES] | [AMFM] | [HUMAN (WAT2022)] | [HUMAN (WAT2021)] | [HUMAN (WAT2020)] | [HUMAN (WAT2019)] | [HUMAN (WAT2018)] | [HUMAN (WAT2017)] | [HUMAN (WAT2016)] | [HUMAN (WAT2015)] | [HUMAN (WAT2014)] | [EVALUATION RESULTS USAGE POLICY]

BLEU


# Team Task Date/Time DataID BLEU
Method
Other
Resources
System
Description
juman kytea mecab moses-
tokenizer
stanford-
segmenter-
ctb
stanford-
segmenter-
pku
indic-
tokenizer
unuse myseg kmseg
1ORGANIZERJPCNja-en2018/08/16 16:54:322000---39.39---- 0.00 0.00NMTNoNMT with Attention
2sarahJPCNja-en2019/07/22 15:01:372805---41.74------NMTNoTransformer, single model
3sarahJPCNja-en2019/07/25 13:07:532927---43.34------NMTNoTransformer, ensemble of 4 models
4ryanJPCNja-en2019/07/26 08:52:532962---42.28------NMTNoBase Transformer
5KNU_HyundaiJPCNja-en2019/07/27 12:24:453188---44.72------NMTYesTransformer Base (+ASPEC data), relative position, BT, r2l reranking, checkpoint ensemble
6goku20JPCNja-en2020/09/18 17:22:333923---43.57------NMTNoTransformer, ensemble of 3 models
7goku20JPCNja-en2020/09/18 17:33:053930---43.51------NMTNomBART pre-training, ensemble of 3 models
8TMUJPCNja-en2021/04/16 21:43:285187---43.78------NMTNoJapanese BART, ensemble of 3 models
9tpt_watJPCNja-en2021/04/27 02:30:535709---41.01------NMTNoBase Transformer model with separate vocab, size 8k
10Bering LabJPCNja-en2021/04/28 13:39:595740---45.13------NMTYesTransformer Ensemble with additional crawled parallel corpus

Notice:

Back to top

RIBES


# Team Task Date/Time DataID RIBES
Method
Other
Resources
System
Description
juman kytea mecab moses-
tokenizer
stanford-
segmenter-
ctb
stanford-
segmenter-
pku
indic-
tokenizer
unuse myseg kmseg
1ORGANIZERJPCNja-en2018/08/16 16:54:322000---0.837932----0.0000000.000000NMTNoNMT with Attention
2sarahJPCNja-en2019/07/22 15:01:372805---0.847227------NMTNoTransformer, single model
3sarahJPCNja-en2019/07/25 13:07:532927---0.853615------NMTNoTransformer, ensemble of 4 models
4ryanJPCNja-en2019/07/26 08:52:532962---0.851879------NMTNoBase Transformer
5KNU_HyundaiJPCNja-en2019/07/27 12:24:453188---0.860706------NMTYesTransformer Base (+ASPEC data), relative position, BT, r2l reranking, checkpoint ensemble
6goku20JPCNja-en2020/09/18 17:22:333923---0.854845------NMTNoTransformer, ensemble of 3 models
7goku20JPCNja-en2020/09/18 17:33:053930---0.857091------NMTNomBART pre-training, ensemble of 3 models
8TMUJPCNja-en2021/04/16 21:43:285187---0.857054------NMTNoJapanese BART, ensemble of 3 models
9tpt_watJPCNja-en2021/04/27 02:30:535709---0.847626------NMTNoBase Transformer model with separate vocab, size 8k
10Bering LabJPCNja-en2021/04/28 13:39:595740---0.866565------NMTYesTransformer Ensemble with additional crawled parallel corpus

Notice:

Back to top

AMFM


# Team Task Date/Time DataID AMFM
Method
Other
Resources
System
Description
unuse unuse unuse unuse unuse unuse unuse unuse unuse unuse
1ORGANIZERJPCNja-en2018/08/16 16:54:322000---0.000000----0.0000000.000000NMTNoNMT with Attention
2sarahJPCNja-en2019/07/22 15:01:372805---0.000000------NMTNoTransformer, single model
3sarahJPCNja-en2019/07/25 13:07:532927---0.000000------NMTNoTransformer, ensemble of 4 models
4ryanJPCNja-en2019/07/26 08:52:532962---0.000000------NMTNoBase Transformer
5KNU_HyundaiJPCNja-en2019/07/27 12:24:453188---0.000000------NMTYesTransformer Base (+ASPEC data), relative position, BT, r2l reranking, checkpoint ensemble
6goku20JPCNja-en2020/09/18 17:22:333923---0.000000------NMTNoTransformer, ensemble of 3 models
7goku20JPCNja-en2020/09/18 17:33:053930---0.000000------NMTNomBART pre-training, ensemble of 3 models
8TMUJPCNja-en2021/04/16 21:43:285187---0.578009------NMTNoJapanese BART, ensemble of 3 models
9tpt_watJPCNja-en2021/04/27 02:30:535709---0.574614------NMTNoBase Transformer model with separate vocab, size 8k
10Bering LabJPCNja-en2021/04/28 13:39:595740---0.579502------NMTYesTransformer Ensemble with additional crawled parallel corpus

Notice:

Back to top

HUMAN (WAT2022)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

HUMAN (WAT2021)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1TMUJPCNja-en2021/04/16 21:43:285187UnderwayNMTNoJapanese BART, ensemble of 3 models

Notice:
Back to top

HUMAN (WAT2020)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1goku20JPCNja-en2020/09/18 17:33:0539304.590NMTNomBART pre-training, ensemble of 3 models

Notice:
Back to top

HUMAN (WAT2019)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1sarahJPCNja-en2019/07/25 13:07:532927UnderwayNMTNoTransformer, ensemble of 4 models
2KNU_HyundaiJPCNja-en2019/07/27 12:24:453188UnderwayNMTYesTransformer Base (+ASPEC data), relative position, BT, r2l reranking, checkpoint ensemble

Notice:
Back to top

HUMAN (WAT2018)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

HUMAN (WAT2017)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

HUMAN (WAT2016)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

HUMAN (WAT2015)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

HUMAN (WAT2014)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description

Notice:
Back to top

EVALUATION RESULTS USAGE POLICY

When you use the WAT evaluation results for any purpose such as:
- writing technical papers,
- making presentations about your system,
- advertising your MT system to the customers,
you can use the information about translation directions, scores (including both automatic and human evaluations) and ranks of your system among others. You can also use the scores of the other systems, but you MUST anonymize the other system's names. In addition, you can show the links (URLs) to the WAT evaluation result pages.

NICT (National Institute of Information and Communications Technology)
Kyoto University
Last Modified: 2018-08-02