Machine Translation on news En-Fi 2015 (dev)
13.72BLEUBase Character-level NMT (Row o)
Evaluation results
| Method | Links | |
|---|---|---|
| Base Character-level NMT (Row o)Target=Char, Depth=2, Attention=Concatenated, Model=Base, Ensemble=true2016.03 | 13.72 | |
| Bi-scale Character-level NMT (Row p)Target=Char, Depth=2, Attention=Concatenated, Model=Bi-S, Ensemble=true2016.03 | 13.39 | |
| Base BPE-level NMT (Row n)Target=BPE, Depth=2, Attention=Concatenated, Model=Base, Ensemble=true2016.03 | 11.92 | |
| Base Character-level NMT (Row o)Target=Char, Depth=2, Attention=Concatenated, Model=Base, Ensemble=false2016.03 | 11.19 | |
| Bi-scale Character-level NMT (Row p)Target=Char, Depth=2, Attention=Concatenated, Model=Bi-S, Ensemble=false2016.03 | 10.73 | |
| Base BPE-level NMT (Row n)Target=BPE, Depth=2, Attention=Concatenated, Model=Base, Ensemble=false2016.03 | 9.61 |