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
1ORGANIZERALT20en-hi2020/09/01 16:01:343611------23.38---NMTNoBaseline MLNMT En to XX model using ALT, Ubuntu, GNOME and KDE4 data from opus. Transformer big model. Default settings.
2NICT-5ALT20en-hi2020/09/18 19:11:433953------24.23---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size balanced.
3NICT-5ALT20en-hi2020/09/18 21:50:484011------22.74---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size unbalanced.
4sakuraALT20en-hi2021/04/29 11:58:055791------34.25---NMTNoMultilingual finetuning of mBART50 finetuned many-to-many model, ensemble of 3
5NICT-2ALT20en-hi2021/05/01 13:15:265908------22.31---NMTNoTransformer base model, multilingual + mixed domain training with domain fine-tuning.
6NICT-2ALT20en-hi2021/05/01 13:26:335916------34.97---NMTYesThe extended mBART model, mixed domain training with domain fine-tuning.
7HwTscSUALT20en-hi2022/07/11 12:03:296736------20.30---NMTNoXX to XX transformer model finetune on the baseline trained on IT domain data

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
1ORGANIZERALT20en-hi2020/09/01 16:01:343611------0.761362---NMTNoBaseline MLNMT En to XX model using ALT, Ubuntu, GNOME and KDE4 data from opus. Transformer big model. Default settings.
2NICT-5ALT20en-hi2020/09/18 19:11:433953------0.772339---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size balanced.
3NICT-5ALT20en-hi2020/09/18 21:50:484011------0.766690---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size unbalanced.
4sakuraALT20en-hi2021/04/29 11:58:055791------0.820590---NMTNoMultilingual finetuning of mBART50 finetuned many-to-many model, ensemble of 3
5NICT-2ALT20en-hi2021/05/01 13:15:265908------0.757269---NMTNoTransformer base model, multilingual + mixed domain training with domain fine-tuning.
6NICT-2ALT20en-hi2021/05/01 13:26:335916------0.822350---NMTYesThe extended mBART model, mixed domain training with domain fine-tuning.
7HwTscSUALT20en-hi2022/07/11 12:03:296736------0.739152---NMTNoXX to XX transformer model finetune on the baseline trained on IT domain data

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
1ORGANIZERALT20en-hi2020/09/01 16:01:343611------0.771225---NMTNoBaseline MLNMT En to XX model using ALT, Ubuntu, GNOME and KDE4 data from opus. Transformer big model. Default settings.
2NICT-5ALT20en-hi2020/09/18 19:11:433953------0.000000---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size balanced.
3NICT-5ALT20en-hi2020/09/18 21:50:484011------0.000000---NMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size unbalanced.
4sakuraALT20en-hi2021/04/29 11:58:055791------0.849202---NMTNoMultilingual finetuning of mBART50 finetuned many-to-many model, ensemble of 3
5NICT-2ALT20en-hi2021/05/01 13:15:265908------0.774041---NMTNoTransformer base model, multilingual + mixed domain training with domain fine-tuning.
6NICT-2ALT20en-hi2021/05/01 13:26:335916------0.839182---NMTYesThe extended mBART model, mixed domain training with domain fine-tuning.
7HwTscSUALT20en-hi2022/07/11 12:03:296736------0.000000---NMTNoXX to XX transformer model finetune on the baseline trained on IT domain data

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
1sakuraALT20en-hi2021/04/29 11:58:055791UnderwayNMTNoMultilingual finetuning of mBART50 finetuned many-to-many model, ensemble of 3
2NICT-2ALT20en-hi2021/05/01 13:26:335916UnderwayNMTYesThe extended mBART model, mixed domain training with domain fine-tuning.

Notice:
Back to top

HUMAN (WAT2020)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1NICT-5ALT20en-hi2020/09/18 19:11:433953UnderwayNMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size balanced.
2NICT-5ALT20en-hi2020/09/18 21:50:484011UnderwayNMTNoXX to XX transformer model trained on ALT as well as KDE, GNOME and Ubuntu data from OPUS. Corpora were size unbalanced.

Notice:
Back to top

HUMAN (WAT2019)


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

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