<?xml version="1.0" standalone="yes"?>
<Paper uid="H89-1036">
  <Title>Literal derivation s NP VP N V NP</Title>
  <Section position="6" start_page="213" end_page="214" type="concl">
    <SectionTitle>
5 Conclusion
</SectionTitle>
    <Paragraph position="0"> We have presented a general parsing strategy based on 'lexicalized' grammars. TAGs are shown to be naturally 'lexicalized'. Lexicalization of a grammar suggests a two-step parsing strategy. The first step selects the set of structures corresponding to each word in the sentence. The second step puts the argument structures into predicate structures. In the first step, structures, rather than non-terminals, are associated with lexical items. The Earley-type parser for TAGs has been extended to take advantage of this strategy.</Paragraph>
    <Paragraph position="1">  We have briefly described the current state of the implementation and the size of the associated lexicon. Finally we show that in lexicalized TAGs idioms can be processed without defining special rules for processing them. We can access simultaneously frozen elements at different levels of depths in contrast to a CFG which either has to flatten the idiomatic structure (and loose the possibility of regular insertion of modifiers) or to use specific devices to check the presence of an idiom. The two pass parsing strategy we use, combined with the operation of direct attachment of lexical items in idiomatic trees, enable us to cut down the number of idiomatic trees that the parser takes as possible candidates. We easily get the possibly idiomatic and literal reading for a given sentence. The only distinctive property of idioms is the non compositional semantics of their frozen constituents.</Paragraph>
  </Section>
class="xml-element"></Paper>