Determining the Placement of German Verbs in English-to-German SMT
Anita Gojun, Alexander Fraser
EACL 2012

When translating English to German, using statistical machine
translation (SMT), existing reordering models often cannot model
long-range reorderings which are needed to generate the German
translations with verbs in the correct position. We implement a method
for reordering English as a preprocessing step for English-to-German
SMT. The reordering is a sequence of hand-crafted reordering rules
applied to English parse trees.  The reordering rules place English
verbal elements in the positions within the clause they will have in
the German translation.  This is a difficult problem, as German verbal
elements can appear in different positions within a clause (in
contrast with English verbal elements, for which the position can not
vary as much).  The reordered English sentences are then translated by
a standard phrase-based SMT system. We show that our approach leads to
a significant improvement in translation performance.