W11-2161 |
Xiaojun Abstract We present the DFKI
|
hybrid translation
|
system at the WMT workshop 2011
|
W11-2161 |
future work We presented the DFKI
|
hybrid translation
|
system submitted in the WMT workshop
|
W09-0411 |
in this paper , we approach a
|
hybrid translation
|
technique combining the output
|
P11-4001 |
5 Conclusion In this paper , a
|
hybrid translation
|
approach for translating the
|
N12-1017 |
this section , we present HyTER (
|
Hybrid Translation
|
Edit Rate ) , a novel metric
|
W12-0111 |
evaluation on test data shows that
|
hybrid translations
|
were generally ranked higher
|
W12-3143 |
data is controlled by the PROMT
|
hybrid translation
|
engine ) . Besides , we have
|
W10-1708 |
systems in all language pairs . The
|
hybrid translation
|
as described in section 2 improves
|
W10-1708 |
steps to cleanup and finalize the
|
hybrid translation
|
result . A schematic overview
|
W12-3138 |
template trans - lations , the
|
hybrid translation
|
using the best candidates as
|
W12-0115 |
RBMT output as template for our
|
hybrid translation
|
approach . The Lucy translation
|
W12-0115 |
potential for more deeply integrated
|
hybrid translation
|
approaches . Nonetheless , the
|
W10-1708 |
of ' Hybrid SG ' shows that the
|
hybrid translation
|
quality can be enhanced if the
|
W12-0111 |
exploits structural output to create
|
hybrid translations
|
and uses the linguistic knowledge
|
W10-1708 |
appropriate " template " for our
|
hybrid translation
|
approach . The Lucy translation
|
W11-2161 |
in the WMT workshop 2011 . The
|
hybrid translation
|
is performed on the final translation
|
W10-1708 |
potential for more deeply integrated
|
hybrid translation
|
approaches . Nevertheless , the
|
W12-0115 |
in section 4.1 . 3.2 Creating
|
Hybrid Translations
|
Using suitable training data
|
W12-4210 |
leads us to the conclusion that a
|
hybrid translation
|
system , a combination of bilingual
|
W10-1708 |
erroneous prepositions are removed . 3
|
Hybrid Translation
|
Analysis We evaluated the intermediate
|