<?xml version="1.0" standalone="yes"?>
<Paper uid="C04-1059">
  <Title>Language Model Adaptation for Statistical Machine Translation with Structured Query Models</Title>
  <Section position="10" start_page="11" end_page="11" type="ackno">
    <SectionTitle>
5 Summary
</SectionTitle>
    <Paragraph position="0"> In this paper, we studied language model adaptation for statistical machine translation.</Paragraph>
    <Paragraph position="1"> Extracting sentences most similar to the initial translations, building specific language models for each sentence to be translated, and interpolating those with the background language models gives significant improvement in translation quality.</Paragraph>
    <Paragraph position="2"> Using structured query models, which capture word order information, leads to better results that plain bag of words models.</Paragraph>
    <Paragraph position="3"> The results obtained suggest a number of extensions of this work: The first question is if more data to retrieve similar sentences from will result in even better translation quality. A second interesting question is if the translation probabilities can be incorporated into the queries. This might be especially useful for structured query models generated from the translation lattices.</Paragraph>
  </Section>
class="xml-element"></Paper>