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
1NICT-5INDICsi-en2018/08/24 14:53:342122---21.85---- 0.00 0.00NMTNoBilingual transformer model
2IITP-MTINDICsi-en2018/09/14 20:00:582352---21.35---- 0.00 0.00NMTNoTransformer multilingual XX-En
3NICT-5INDICsi-en2018/09/07 14:33:362244---19.19---- 0.00 0.00NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
4NICT-5INDICsi-en2018/08/24 14:53:512123---18.73---- 0.00 0.00NMTNoXX-En transformer model
5ORGANIZERINDICsi-en2018/08/20 11:28:472014---18.16---- 0.00 0.00NMTNoNMT with Attention
6ORGANIZERINDICsi-en2018/08/24 14:42:072105---17.41---- 0.00 0.00NMTNomulti2one multilingual NMT with Attention
7NICT-5INDICsi-en2018/08/24 15:01:552139---17.25---- 0.00 0.00NMTNoXX-XX transformer model
8AnuvaadINDICsi-en2018/09/15 18:09:232412---16.92---- 0.00 0.00SMTNoSMT XX-En
9ORGANIZERINDICsi-en2018/08/29 14:38:182199---16.89---- 0.00 0.00NMTNomulti2multi multilingual NMT with Attention
10AnuvaadINDICsi-en2018/09/15 18:34:312413---16.44---- 0.00 0.00SMTNoSMT with KenLM
11RGNLPINDICsi-en2018/09/15 03:22:252388---13.21---- 0.00 0.00NMTNoNMT system with a 2-layer LSTM method
12RGNLPINDICsi-en2018/09/15 02:37:272369---13.10---- 0.00 0.00SMTNoSMT system with KENLM Language model
13RGNLPINDICsi-en2018/09/15 02:53:132378---12.79---- 0.00 0.00SMTNoSMT system with SRILM Language model

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
1NICT-5INDICsi-en2018/08/24 14:53:342122---0.766387----0.0000000.000000NMTNoBilingual transformer model
2IITP-MTINDICsi-en2018/09/14 20:00:582352---0.760533----0.0000000.000000NMTNoTransformer multilingual XX-En
3NICT-5INDICsi-en2018/09/07 14:33:362244---0.756363----0.0000000.000000NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
4NICT-5INDICsi-en2018/08/24 14:53:512123---0.754882----0.0000000.000000NMTNoXX-En transformer model
5NICT-5INDICsi-en2018/08/24 15:01:552139---0.746178----0.0000000.000000NMTNoXX-XX transformer model
6ORGANIZERINDICsi-en2018/08/24 14:42:072105---0.744761----0.0000000.000000NMTNomulti2one multilingual NMT with Attention
7ORGANIZERINDICsi-en2018/08/29 14:38:182199---0.742163----0.0000000.000000NMTNomulti2multi multilingual NMT with Attention
8ORGANIZERINDICsi-en2018/08/20 11:28:472014---0.739595----0.0000000.000000NMTNoNMT with Attention
9RGNLPINDICsi-en2018/09/15 03:22:252388---0.707741----0.0000000.000000NMTNoNMT system with a 2-layer LSTM method
10AnuvaadINDICsi-en2018/09/15 18:34:312413---0.692275----0.0000000.000000SMTNoSMT with KenLM
11AnuvaadINDICsi-en2018/09/15 18:09:232412---0.692236----0.0000000.000000SMTNoSMT XX-En
12RGNLPINDICsi-en2018/09/15 02:37:272369---0.661738----0.0000000.000000SMTNoSMT system with KENLM Language model
13RGNLPINDICsi-en2018/09/15 02:53:132378---0.658720----0.0000000.000000SMTNoSMT system with SRILM 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
1IITP-MTINDICsi-en2018/09/14 20:00:582352---0.551630----0.0000000.000000NMTNoTransformer multilingual XX-En
2NICT-5INDICsi-en2018/08/24 14:53:342122---0.534840----0.0000000.000000NMTNoBilingual transformer model
3NICT-5INDICsi-en2018/08/24 14:53:512123---0.528530----0.0000000.000000NMTNoXX-En transformer model
4NICT-5INDICsi-en2018/09/07 14:33:362244---0.526570----0.0000000.000000NMTNoUnified Source vocabulary by orthography mapping. XX-EN model.
5NICT-5INDICsi-en2018/08/24 15:01:552139---0.512610----0.0000000.000000NMTNoXX-XX transformer model
6ORGANIZERINDICsi-en2018/08/20 11:28:472014---0.507670----0.0000000.000000NMTNoNMT with Attention
7ORGANIZERINDICsi-en2018/08/24 14:42:072105---0.506660----0.0000000.000000NMTNomulti2one multilingual NMT with Attention
8ORGANIZERINDICsi-en2018/08/29 14:38:182199---0.505990----0.0000000.000000NMTNomulti2multi multilingual NMT with Attention
9AnuvaadINDICsi-en2018/09/15 18:34:312413---0.492410----0.0000000.000000SMTNoSMT with KenLM
10RGNLPINDICsi-en2018/09/15 02:53:132378---0.488230----0.0000000.000000SMTNoSMT system with SRILM Language model
11RGNLPINDICsi-en2018/09/15 03:22:252388---0.486370----0.0000000.000000NMTNoNMT system with a 2-layer LSTM method
12AnuvaadINDICsi-en2018/09/15 18:09:232412---0.484710----0.0000000.000000SMTNoSMT XX-En
13RGNLPINDICsi-en2018/09/15 02:37:272369---0.484230----0.0000000.000000SMTNoSMT system with KENLM Language model

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