NICT_LOGO.JPG KYOTO-U_LOGO.JPG

WAT

The Workshop on Asian Translation
Evaluation Results

[EVALUATION RESULTS TOP] | [BLEU] | [RIBES] | [AMFM] | [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
1FBAIALT2my-en2019/07/27 14:36:583201---38.59------NMTYes5 model ensemble, (BT iter1 + Self-Training) + (BT+ST) iter2 + fine tuning + noisy channel
2NICTALT2my-en2019/07/22 17:06:592816---30.15------NMTYesSingle model+language model pre-training+back-translation
3FBAIALT2my-en2019/07/27 08:05:363148---26.75------NMTNoensemble 5 models, (BT iter1 + Self-Training)
4NICT-4ALT2my-en2019/07/26 11:31:522977---24.75------OtherYesSame as last year but with cleaner monolingual data
5UCSYNLPALT2my-en2019/07/29 13:51:403252---19.64------NMTNoNMT with Attention
6NICTALT2my-en2019/07/23 13:40:552854---18.51------NMTNoSingle model+language model pre-training
7ORGANIZERALT2my-en2019/07/22 19:08:032826---14.85------NMTNoNMT with Attention
8ORGANIZERALT2my-en2019/07/24 16:17:222899---14.59------OtherYesOnline A
9UCSMNLPALT2my-en2019/07/26 17:01:163022---10.70------SMTYes

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
1FBAIALT2my-en2019/07/27 14:36:583201---0.840001------NMTYes5 model ensemble, (BT iter1 + Self-Training) + (BT+ST) iter2 + fine tuning + noisy channel
2NICTALT2my-en2019/07/22 17:06:592816---0.791705------NMTYesSingle model+language model pre-training+back-translation
3FBAIALT2my-en2019/07/27 08:05:363148---0.783571------NMTNoensemble 5 models, (BT iter1 + Self-Training)
4NICT-4ALT2my-en2019/07/26 11:31:522977---0.760394------OtherYesSame as last year but with cleaner monolingual data
5NICTALT2my-en2019/07/23 13:40:552854---0.744808------NMTNoSingle model+language model pre-training
6UCSYNLPALT2my-en2019/07/29 13:51:403252---0.707789------NMTNoNMT with Attention
7ORGANIZERALT2my-en2019/07/22 19:08:032826---0.700166------NMTNoNMT with Attention
8ORGANIZERALT2my-en2019/07/24 16:17:222899---0.602267------OtherYesOnline A
9UCSMNLPALT2my-en2019/07/26 17:01:163022---0.570835------SMTYes

Notice:

Back to top

AMFM


# Team Task Date/Time DataID AMFM
Method
Other
Resources
System
Description
juman kytea mecab moses-
tokenizer
stanford-
segmenter-
ctb
stanford-
segmenter-
pku
indic-
tokenizer
unuse myseg kmseg
1FBAIALT2my-en2019/07/27 14:36:583201---0.685200------NMTYes5 model ensemble, (BT iter1 + Self-Training) + (BT+ST) iter2 + fine tuning + noisy channel
2NICTALT2my-en2019/07/22 17:06:592816---0.649050------NMTYesSingle model+language model pre-training+back-translation
3FBAIALT2my-en2019/07/27 08:05:363148---0.627530------NMTNoensemble 5 models, (BT iter1 + Self-Training)
4NICT-4ALT2my-en2019/07/26 11:31:522977---0.579570------OtherYesSame as last year but with cleaner monolingual data
5NICTALT2my-en2019/07/23 13:40:552854---0.565430------NMTNoSingle model+language model pre-training
6ORGANIZERALT2my-en2019/07/24 16:17:222899---0.549380------OtherYesOnline A
7UCSMNLPALT2my-en2019/07/26 17:01:163022---0.538280------SMTYes
8UCSYNLPALT2my-en2019/07/29 13:51:403252---0.532640------NMTNoNMT with Attention
9ORGANIZERALT2my-en2019/07/22 19:08:032826---0.467460------NMTNoNMT with Attention

Notice:

Back to top

HUMAN (WAT2019)


# Team Task Date/Time DataID HUMAN
Method
Other
Resources
System
Description
1NICTALT2my-en2019/07/22 17:06:592816UnderwayNMTYesSingle model+language model pre-training+back-translation
2NICT-4ALT2my-en2019/07/26 11:31:522977UnderwayOtherYesSame as last year but with cleaner monolingual data
3UCSMNLPALT2my-en2019/07/26 17:01:163022UnderwaySMTYes
4FBAIALT2my-en2019/07/27 08:05:363148UnderwayNMTNoensemble 5 models, (BT iter1 + Self-Training)
5FBAIALT2my-en2019/07/27 14:36:583201UnderwayNMTYes5 model ensemble, (BT iter1 + Self-Training) + (BT+ST) iter2 + fine tuning + noisy channel
6UCSYNLPALT2my-en2019/07/29 13:51:403252UnderwayNMTNoNMT with Attention

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