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
1Bering LabJPCNja-zh2021/04/30 14:57:345854-42.92--43.9043.61----NMTYesTransformer Ensemble with additional crawled parallel corpus
2KNU_HyundaiJPCNja-zh2019/07/27 08:32:593157-42.75--43.8843.30----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
3goku20JPCNja-zh2020/09/18 17:13:293917-41.65--42.7442.51----NMTNoTransformer, ensemble of 3 models
4goku20JPCNja-zh2020/09/18 17:26:253926-41.52--42.5942.27----NMTNomBART pre-training, ensemble of 3 models
5sarahJPCNja-zh2019/07/25 12:20:442921-41.45--42.4342.22----NMTNoTransformer, ensemble of 4 models
6ryanJPCNja-zh2019/07/25 21:55:312948-40.62--41.6041.25----NMTNoBase Transformer
7sarahJPCNja-zh2019/07/22 14:48:472803-39.98--41.0040.78----NMTNoTransformer, single model
8USTCJPCNja-zh2018/08/31 16:48:482201-38.96--39.6439.34-- 0.00 0.00NMTNotensor2tensor, 4 model average, r2l rerank
9ORGANIZERJPCNja-zh2018/08/16 16:18:181996-38.07--39.1838.65-- 0.00 0.00NMTNoNMT with Attention
10tpt_watJPCNja-zh2021/04/27 01:46:025693-37.27--38.3038.04----NMTNoBase Transformer model with shared vocab 8k size

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
1Bering LabJPCNja-zh2021/04/30 14:57:345854-0.867426--0.8716430.870725----NMTYesTransformer Ensemble with additional crawled parallel corpus
2KNU_HyundaiJPCNja-zh2019/07/27 08:32:593157-0.865470--0.8700930.869199----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
3ryanJPCNja-zh2019/07/25 21:55:312948-0.854858--0.8596020.858790----NMTNoBase Transformer
4sarahJPCNja-zh2019/07/25 12:20:442921-0.853633--0.8588230.857595----NMTNoTransformer, ensemble of 4 models
5goku20JPCNja-zh2020/09/18 17:26:253926-0.853135--0.8591180.857898----NMTNomBART pre-training, ensemble of 3 models
6goku20JPCNja-zh2020/09/18 17:13:293917-0.852752--0.8580600.857049----NMTNoTransformer, ensemble of 3 models
7USTCJPCNja-zh2018/08/31 16:48:482201-0.849951--0.8546790.853397--0.0000000.000000NMTNotensor2tensor, 4 model average, r2l rerank
8sarahJPCNja-zh2019/07/22 14:48:472803-0.848909--0.8543140.852927----NMTNoTransformer, single model
9ORGANIZERJPCNja-zh2018/08/16 16:18:181996-0.845075--0.8491180.848360--0.0000000.000000NMTNoNMT with Attention
10tpt_watJPCNja-zh2021/04/27 01:46:025693-0.841730--0.8462430.845044----NMTNoBase Transformer model with shared vocab 8k size

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
1Bering LabJPCNja-zh2021/04/30 14:57:345854-0.895420--0.8954200.895420----NMTYesTransformer Ensemble with additional crawled parallel corpus
2tpt_watJPCNja-zh2021/04/27 01:46:025693-0.885424--0.8854240.885424----NMTNoBase Transformer model with shared vocab 8k size
3ORGANIZERJPCNja-zh2018/08/16 16:18:181996-0.000000--0.0000000.000000--0.0000000.000000NMTNoNMT with Attention
4USTCJPCNja-zh2018/08/31 16:48:482201-0.000000--0.0000000.000000--0.0000000.000000NMTNotensor2tensor, 4 model average, r2l rerank
5sarahJPCNja-zh2019/07/22 14:48:472803-0.000000--0.0000000.000000----NMTNoTransformer, single model
6sarahJPCNja-zh2019/07/25 12:20:442921-0.000000--0.0000000.000000----NMTNoTransformer, ensemble of 4 models
7ryanJPCNja-zh2019/07/25 21:55:312948-0.000000--0.0000000.000000----NMTNoBase Transformer
8KNU_HyundaiJPCNja-zh2019/07/27 08:32:593157-0.000000--0.0000000.000000----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
9goku20JPCNja-zh2020/09/18 17:13:293917-0.000000--0.0000000.000000----NMTNoTransformer, ensemble of 3 models
10goku20JPCNja-zh2020/09/18 17:26:253926-0.000000--0.0000000.000000----NMTNomBART pre-training, ensemble of 3 models

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

Notice:
Back to top

HUMAN (WAT2020)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1goku20JPCNja-zh2020/09/18 17:26:2539264.640NMTNomBART pre-training, ensemble of 3 models

Notice:
Back to top

HUMAN (WAT2019)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1sarahJPCNja-zh2019/07/25 12:20:442921UnderwayNMTNoTransformer, ensemble of 4 models
2KNU_HyundaiJPCNja-zh2019/07/27 08:32:593157UnderwayNMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)

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