<?xml version="1.0" standalone="yes"?>
<Paper uid="W04-0902">
  <Title>Solving Logic Puzzles: From Robust Processing to Precise Semantics</Title>
  <Section position="12" start_page="0" end_page="0" type="concl">
    <SectionTitle>
11 Conclusion and Further Work
</SectionTitle>
    <Paragraph position="0"> The key open problem is identifying sufficiently robust and general methods for building precise semantic representations, rather than requiring hand-built translation rules for a seemingly endless list of special phenomena. Immediate future work will include extending and generalizing the system's coverage of syntax-to-semantics mappings, incorporating classifiers for suggesting likely coreference resolutions and operator scopings, and developing methods for calculating presuppositions and inferences. This work may be sufficient to give good coverage of the problem domain, or we may need to develop new more robust models of syntactic to semantic transductions.</Paragraph>
  </Section>
class="xml-element"></Paper>