Table 3: 
Test performance. NC-v11 represents training only with the NC-v11 data, while Full means using the full training data. * represents significant (Koehn, 2004) result (p < 0.01) over Seq2seq. ↓ indicates the lower the better.
SystemNC-v11Full
BLEUTER↓MeteorBLEUTER↓Meteor
OpenNMT-tf 15.1 0.6902 0.3040 24.3 0.5567 0.4225 
Transformer-tf 17.1 0.6647 0.3578 25.1 0.5537 0.4344 
Seq2seq 16.0 0.6695 0.3379 23.7 0.5590 0.4258 
Dual2seq-LinAMR 17.3 0.6530 0.3612 24.0 0.5643 0.4246 
Duel2seq-SRL 17.2 0.6591 0.3644 23.8 0.5626 0.4223 
Dual2seq-Dep 17.8 0.6516 0.3673 25.0 0.5538 0.4328 
Dual2seq 19.2* 0.6305 0.3840 25.5* 0.5480 0.4376 
SystemNC-v11Full
BLEUTER↓MeteorBLEUTER↓Meteor
OpenNMT-tf 15.1 0.6902 0.3040 24.3 0.5567 0.4225 
Transformer-tf 17.1 0.6647 0.3578 25.1 0.5537 0.4344 
Seq2seq 16.0 0.6695 0.3379 23.7 0.5590 0.4258 
Dual2seq-LinAMR 17.3 0.6530 0.3612 24.0 0.5643 0.4246 
Duel2seq-SRL 17.2 0.6591 0.3644 23.8 0.5626 0.4223 
Dual2seq-Dep 17.8 0.6516 0.3673 25.0 0.5538 0.4328 
Dual2seq 19.2* 0.6305 0.3840 25.5* 0.5480 0.4376 
View Large
Close Modal
This Feature Is Available To Subscribers Only

or Create an Account

Close Modal
Close Modal
This site uses cookies. By continuing to use our website, you are agreeing to our privacy policy. No content on this site may be used to train artificial intelligence systems without permission in writing from the MIT Press.
Accept