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
1ORGANIZERJPCN2ja-zh2018/08/15 18:23:171961-39.14--40.2839.77-- 0.00 0.00NMTNoNMT with Attention
2USTCJPCN2ja-zh2018/08/31 17:00:532203-39.91--40.5340.32-- 0.00 0.00NMTNotensor2tensor, 4 model average, r2l rerank
3ryanJPCN2ja-zh2019/07/25 22:06:172951-41.68--42.6342.26----NMTNoBase Transformer
4sarahJPCN2ja-zh2019/07/26 11:40:152983-42.23--43.1942.98----NMTNoTransformer, ensemble of 4 models
5KNU_HyundaiJPCN2ja-zh2019/07/27 08:39:213165-44.09--45.2344.68----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
6goku20JPCN2ja-zh2020/09/21 12:09:424090-40.84--41.7541.43----NMTNomBART pre-training transformer, single model
7goku20JPCN2ja-zh2020/09/22 00:10:384110-41.93--42.8442.52----NMTNomBART pre-training transformer, ensemble of 3 models
8goku20JPCN2ja-zh2020/09/22 00:12:524111-41.93--42.8442.52----NMTNomBART pre-training transformer, ensemble of 3 models
9tpt_watJPCN2ja-zh2021/04/27 01:50:285695-38.76--39.7139.47----SMTNoBase Transformer Model with separate vocabularies, 8k size
10tpt_watJPCN2ja-zh2021/04/27 01:51:045696-38.76--39.7139.47----SMTNoBase Transformer Model with separate vocabularies, 8k size
11Bering LabJPCN2ja-zh2021/04/30 12:39:495848-43.98--45.0144.72----NMTYesTransformer Ensemble with additional crawled parallel corpus
12sakuraJPCN2ja-zh2024/08/09 00:42:327304-40.60--41.4041.10----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh) - Best
13sakuraJPCN2ja-zh2024/08/09 00:44:497305-40.30--41.2040.80----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh)

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
1ORGANIZERJPCN2ja-zh2018/08/15 18:23:171961-0.847486--0.8521610.850707--0.0000000.000000NMTNoNMT with Attention
2USTCJPCN2ja-zh2018/08/31 17:00:532203-0.853978--0.8593300.857456--0.0000000.000000NMTNotensor2tensor, 4 model average, r2l rerank
3ryanJPCN2ja-zh2019/07/25 22:06:172951-0.857546--0.8625990.861150----NMTNoBase Transformer
4sarahJPCN2ja-zh2019/07/26 11:40:152983-0.854775--0.8608370.859031----NMTNoTransformer, ensemble of 4 models
5KNU_HyundaiJPCN2ja-zh2019/07/27 08:39:213165-0.869556--0.8741400.872997----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
6goku20JPCN2ja-zh2020/09/21 12:09:424090-0.853709--0.8598720.858557----NMTNomBART pre-training transformer, single model
7goku20JPCN2ja-zh2020/09/22 00:10:384110-0.857293--0.8634190.861801----NMTNomBART pre-training transformer, ensemble of 3 models
8goku20JPCN2ja-zh2020/09/22 00:12:524111-0.857293--0.8634190.861801----NMTNomBART pre-training transformer, ensemble of 3 models
9tpt_watJPCN2ja-zh2021/04/27 01:50:285695-0.844570--0.8498190.848185----SMTNoBase Transformer Model with separate vocabularies, 8k size
10tpt_watJPCN2ja-zh2021/04/27 01:51:045696-0.844570--0.8498190.848185----SMTNoBase Transformer Model with separate vocabularies, 8k size
11Bering LabJPCN2ja-zh2021/04/30 12:39:495848-0.868941--0.8733820.872298----NMTYesTransformer Ensemble with additional crawled parallel corpus
12sakuraJPCN2ja-zh2024/08/09 00:42:327304-0.854800--0.8608330.859124----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh) - Best
13sakuraJPCN2ja-zh2024/08/09 00:44:497305-0.854198--0.8601370.858608----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh)

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
1ORGANIZERJPCN2ja-zh2018/08/15 18:23:171961-0.000000--0.0000000.000000--0.0000000.000000NMTNoNMT with Attention
2USTCJPCN2ja-zh2018/08/31 17:00:532203-0.000000--0.0000000.000000--0.0000000.000000NMTNotensor2tensor, 4 model average, r2l rerank
3ryanJPCN2ja-zh2019/07/25 22:06:172951-0.000000--0.0000000.000000----NMTNoBase Transformer
4sarahJPCN2ja-zh2019/07/26 11:40:152983-0.000000--0.0000000.000000----NMTNoTransformer, ensemble of 4 models
5KNU_HyundaiJPCN2ja-zh2019/07/27 08:39:213165-0.000000--0.0000000.000000----NMTYesTransformer(base) + *Used ASPEC corpus* with relative position, bt, r2l rerank, 4-model ensemble(9-check point)
6goku20JPCN2ja-zh2020/09/21 12:09:424090-0.000000--0.0000000.000000----NMTNomBART pre-training transformer, single model
7goku20JPCN2ja-zh2020/09/22 00:10:384110-0.000000--0.0000000.000000----NMTNomBART pre-training transformer, ensemble of 3 models
8goku20JPCN2ja-zh2020/09/22 00:12:524111-0.000000--0.0000000.000000----NMTNomBART pre-training transformer, ensemble of 3 models
9tpt_watJPCN2ja-zh2021/04/27 01:50:285695-0.888029--0.8880290.888029----SMTNoBase Transformer Model with separate vocabularies, 8k size
10tpt_watJPCN2ja-zh2021/04/27 01:51:045696-0.888029--0.8880290.888029----SMTNoBase Transformer Model with separate vocabularies, 8k size
11Bering LabJPCN2ja-zh2021/04/30 12:39:495848-0.893837--0.8938370.893837----NMTYesTransformer Ensemble with additional crawled parallel corpus
12sakuraJPCN2ja-zh2024/08/09 00:42:327304-0.000000--0.0000000.000000----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh) - Best
13sakuraJPCN2ja-zh2024/08/09 00:44:497305-0.000000--0.0000000.000000----NMTNoLLM: Rakuten/RakutenAI-7B-chat Fine-Tuned with JPC Corpus in six direction (En-Ja, Ja-En, Ko-Ja, Ja-Ko, Zh-Ja, Ja-Zh)

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

Notice:
Back to top

HUMAN (WAT2019)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1sarahJPCN2ja-zh2019/07/26 11:40:152983UnderwayNMTNoTransformer, ensemble of 4 models
2KNU_HyundaiJPCN2ja-zh2019/07/27 08:39:213165UnderwayNMTYesTransformer(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