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
1IITP-MTINDIChi-en2018/09/14 19:49:212347---32.95---- 0.00 0.00NMTNoTransformer multilingual XX-En
2cvitINDIChi-en2019/03/23 12:38:282661---32.79---- 0.00 0.00NMTYesmassive-multi e270
3cvitINDIChi-en2019/03/14 22:01:292624---31.55---- 0.00 0.00NMTYesmassive-multi + ft
4NICT-5INDIChi-en2018/09/07 14:29:052238---31.51---- 0.00 0.00NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
5NICT-5INDIChi-en2018/08/22 19:07:052066---31.06---- 0.00 0.00NMTNoXX-En transformer model
6NICT-5INDIChi-en2018/08/24 14:59:122129---30.21---- 0.00 0.00NMTNoXX-XX transformer model
7cvitINDIChi-en2019/03/14 21:49:192616---29.74---- 0.00 0.00NMTYesmassive-multi
8cvitINDIChi-en2019/03/22 05:38:522652---29.00---- 0.00 0.00NMTNomany to en (Transformer)
9cvitINDIChi-en2019/03/22 05:28:122650---28.52---- 0.00 0.00NMTYesmany to en (Transformer)
10ORGANIZERINDIChi-en2018/08/24 14:36:082099---26.71---- 0.00 0.00NMTNomulti2one multilingual NMT with Attention
11ORGANIZERINDIChi-en2018/08/29 14:24:102189---26.55---- 0.00 0.00NMTNomulti2multi multilingual NMT with Attention
12NICT-5INDIChi-en2018/08/22 19:06:452065---26.05---- 0.00 0.00NMTNoBilingual transformer model
13AnuvaadINDIChi-en2018/09/15 17:55:582403---25.57---- 0.00 0.00SMTNoSMT with KenLM
14AnuvaadINDIChi-en2018/09/15 18:01:052406---22.45---- 0.00 0.00SMTNoSMT XX-En
15RGNLPINDIChi-en2018/09/15 03:16:012383---21.86---- 0.00 0.00NMTNoNMT system with a 2-layer LSTM method
16RGNLPINDIChi-en2018/09/15 02:27:032367---21.54---- 0.00 0.00SMTNoSMT system with KENLM Language model
17RGNLPINDIChi-en2018/09/15 02:48:342374---21.16---- 0.00 0.00SMTNoSMT system with SRILM Language model
18ORGANIZERINDIChi-en2018/08/20 11:11:432004---21.15---- 0.00 0.00NMTNoNMT with Attention

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
1cvitINDIChi-en2019/03/23 12:38:282661---0.809092----0.0000000.000000NMTYesmassive-multi e270
2cvitINDIChi-en2019/03/14 22:01:292624---0.805115----0.0000000.000000NMTYesmassive-multi + ft
3IITP-MTINDIChi-en2018/09/14 19:49:212347---0.803497----0.0000000.000000NMTNoTransformer multilingual XX-En
4NICT-5INDIChi-en2018/08/24 14:59:122129---0.801291----0.0000000.000000NMTNoXX-XX transformer model
5cvitINDIChi-en2019/03/22 05:38:522652---0.800158----0.0000000.000000NMTNomany to en (Transformer)
6cvitINDIChi-en2019/03/14 21:49:192616---0.794221----0.0000000.000000NMTYesmassive-multi
7cvitINDIChi-en2019/03/22 05:28:122650---0.793866----0.0000000.000000NMTYesmany to en (Transformer)
8ORGANIZERINDIChi-en2018/08/24 14:36:082099---0.787645----0.0000000.000000NMTNomulti2one multilingual NMT with Attention
9NICT-5INDIChi-en2018/08/22 19:07:052066---0.786655----0.0000000.000000NMTNoXX-En transformer model
10ORGANIZERINDIChi-en2018/08/29 14:24:102189---0.784968----0.0000000.000000NMTNomulti2multi multilingual NMT with Attention
11NICT-5INDIChi-en2018/09/07 14:29:052238---0.784228----0.0000000.000000NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
12NICT-5INDIChi-en2018/08/22 19:06:452065---0.762405----0.0000000.000000NMTNoBilingual transformer model
13ORGANIZERINDIChi-en2018/08/20 11:11:432004---0.752783----0.0000000.000000NMTNoNMT with Attention
14RGNLPINDIChi-en2018/09/15 03:16:012383---0.751517----0.0000000.000000NMTNoNMT system with a 2-layer LSTM method
15AnuvaadINDIChi-en2018/09/15 17:55:582403---0.720866----0.0000000.000000SMTNoSMT with KenLM
16AnuvaadINDIChi-en2018/09/15 18:01:052406---0.709235----0.0000000.000000SMTNoSMT XX-En
17RGNLPINDIChi-en2018/09/15 02:48:342374---0.698427----0.0000000.000000SMTNoSMT system with SRILM Language model
18RGNLPINDIChi-en2018/09/15 02:27:032367---0.697379----0.0000000.000000SMTNoSMT system with KENLM Language model

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
1cvitINDIChi-en2019/03/23 12:38:282661---0.641340----0.0000000.000000NMTYesmassive-multi e270
2cvitINDIChi-en2019/03/14 22:01:292624---0.633010----0.0000000.000000NMTYesmassive-multi + ft
3IITP-MTINDIChi-en2018/09/14 19:49:212347---0.629890----0.0000000.000000NMTNoTransformer multilingual XX-En
4cvitINDIChi-en2019/03/14 21:49:192616---0.622810----0.0000000.000000NMTYesmassive-multi
5NICT-5INDIChi-en2018/08/24 14:59:122129---0.617860----0.0000000.000000NMTNoXX-XX transformer model
6cvitINDIChi-en2019/03/22 05:28:122650---0.614880----0.0000000.000000NMTYesmany to en (Transformer)
7cvitINDIChi-en2019/03/22 05:38:522652---0.614880----0.0000000.000000NMTNomany to en (Transformer)
8AnuvaadINDIChi-en2018/09/15 17:55:582403---0.599880----0.0000000.000000SMTNoSMT with KenLM
9RGNLPINDIChi-en2018/09/15 02:27:032367---0.599760----0.0000000.000000SMTNoSMT system with KENLM Language model
10NICT-5INDIChi-en2018/08/22 19:07:052066---0.599000----0.0000000.000000NMTNoXX-En transformer model
11RGNLPINDIChi-en2018/09/15 02:48:342374---0.597650----0.0000000.000000SMTNoSMT system with SRILM Language model
12NICT-5INDIChi-en2018/09/07 14:29:052238---0.595500----0.0000000.000000NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
13ORGANIZERINDIChi-en2018/08/24 14:36:082099---0.586760----0.0000000.000000NMTNomulti2one multilingual NMT with Attention
14ORGANIZERINDIChi-en2018/08/29 14:24:102189---0.577010----0.0000000.000000NMTNomulti2multi multilingual NMT with Attention
15NICT-5INDIChi-en2018/08/22 19:06:452065---0.574320----0.0000000.000000NMTNoBilingual transformer model
16RGNLPINDIChi-en2018/09/15 03:16:012383---0.573160----0.0000000.000000NMTNoNMT system with a 2-layer LSTM method
17ORGANIZERINDIChi-en2018/08/20 11:11:432004---0.559420----0.0000000.000000NMTNoNMT with Attention
18AnuvaadINDIChi-en2018/09/15 18:01:052406---0.558850----0.0000000.000000SMTNoSMT XX-En

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
1IITP-MTINDIChi-en2018/09/14 19:49:21234740.750NMTNoTransformer multilingual XX-En
2NICT-5INDIChi-en2018/08/24 14:59:12212932.000NMTNoXX-XX transformer model
3RGNLPINDIChi-en2018/09/15 02:27:03236722.250SMTNoSMT system with KENLM Language model
4NICT-5INDIChi-en2018/08/22 19:07:05206613.500NMTNoXX-En transformer model
5AnuvaadINDIChi-en2018/09/15 18:01:0524065.750SMTNoSMT XX-En
6AnuvaadINDIChi-en2018/09/15 17:55:5824030.750SMTNoSMT with KenLM
7RGNLPINDIChi-en2018/09/15 03:16:012383-1.250NMTNoNMT system with a 2-layer LSTM method

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