We directly investigate a subject of much recent debate : do
<term>
word sense disambigation models
</term>
help
<term>
statistical machine translation quality
</term>
?
#9179We directly investigate a subject of much recent debate: doword sense disambigation models help statistical machine translation quality?
measure(ment),16-1-P05-1048,ak
We directly investigate a subject of much recent debate : do
<term>
word sense disambigation models
</term>
help
<term>
statistical machine translation quality
</term>
?
#9184We directly investigate a subject of much recent debate: do word sense disambigation models helpstatistical machine translation quality?
model,3-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9207Using a state-of-the-artChinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone.
other,10-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9214Using a state-of-the-art Chinese word sense disambiguation model to choosetranslation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone.
tech,15-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9219Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typicalIBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone.
tech,23-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9227Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find thatword sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone.
measure(ment),31-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9235Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly bettertranslation quality than the statistical machine translation system alone.
tech,35-3-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
#9239Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than thestatistical machine translation system alone.
tech,0-4-P05-1048,ak
Using a state-of-the-art
<term>
Chinese word sense disambiguation model
</term>
to choose
<term>
translation candidates
</term>
for a typical
<term>
IBM statistical MT system
</term>
, we find that
<term>
word sense disambiguation
</term>
does not yield significantly better
<term>
translation quality
</term>
than the
<term>
statistical machine translation system
</term>
alone .
<term>
Error analysis
</term>
suggests several key factors behind this surprising finding , including inherent limitations of current
<term>
statistical MT architectures
</term>
.
#9245Using a state-of-the-art Chinese word sense disambiguation model to choose translation candidates for a typical IBM statistical MT system, we find that word sense disambiguation does not yield significantly better translation quality than the statistical machine translation system alone.Error analysis suggests several key factors behind this surprising finding, including inherent limitations of current statistical MT architectures.
other,16-4-P05-1048,ak
<term>
Error analysis
</term>
suggests several key factors behind this surprising finding , including inherent limitations of current
<term>
statistical MT architectures
</term>
.
#9261Error analysis suggests several key factors behind this surprising finding, including inherent limitations of currentstatistical MT architectures.