<?xml version="1.0" standalone="yes"?>
<Paper uid="W02-1506">
  <Title>Adapting Existing Grammars: The XLE Experience</Title>
  <Section position="13" start_page="0" end_page="0" type="ackno">
    <SectionTitle>
STANDARD EUREKA WSJ
</SectionTitle>
    <Paragraph position="0"> states 4,935 5,132 8,759 arcs 13,268 13,639 19,695 In sum, the grammar specialization system used in XLE has been quite sucessful in developing corpus specific grammars using the STANDARD English grammar as a basis. A significant benefit comes from being able to distinguish truly unusual constructions that exist only in the specialized grammar from those that are (or should be) in the STANDARD grammar.</Paragraph>
    <Paragraph position="1"> This allows idiosyncratic information to remain in a specialized grammar while all the specialized grammars benefit from and contribute to the continuing development of the STANDARD grammar.</Paragraph>
  </Section>
class="xml-element"></Paper>