other,12-2-N04-1022,bq |
of translation errors
</term>
under
<term>
|
loss functions
|
</term>
that measure
<term>
translation performance
|
#6568
This statistical approach aims to minimize expected loss of translation errors under loss functions that measure translation performance. |
other,5-3-N04-1022,bq |
</term>
. We describe a hierarchy of
<term>
|
loss functions
|
</term>
that incorporate different levels
|
#6580
We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings, word-to-word alignments from an MT system, and syntactic structure from parse-trees of source and target language sentences. |
other,15-3-N04-1022,bq |
linguistic information
</term>
from
<term>
|
word strings
|
</term>
,
<term>
word-to-word alignments
</term>
|
#6590
We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings , word-to-word alignments from an MT system, and syntactic structure from parse-trees of source and target language sentences. |
other,16-5-N04-1022,bq |
MT
</term>
performance for specific
<term>
|
loss functions
|
</term>
.
<term>
CriterionSM Online Essay
|
#6642
Our results show that MBR decoding can be used to tune statistical MT performance for specific loss functions . |