<?xml version="1.0" standalone="yes"?>
<Paper uid="P99-1059">
  <Title>Efficient Parsing for Bilexical Context-Free Grammars and Head Automaton Grammars*</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> Lexicalized grammar formalisms are of both theoretical and practical interest to the computational linguistics community. Such formalisms specify syntactic facts about each word of the language--in particular, the type of arguments that the word can or must take.</Paragraph>
    <Paragraph position="1"> Early mechanisms of this sort included categorial grammar (Bar-Hillel, 1953) and subcategorization frames (Chomsky, 1965). Other lexicalized formalisms include (Schabes et al., 1988; Mel'~uk, 1988; Pollard and Sag, 1994).</Paragraph>
    <Paragraph position="2"> Besides the possible arguments of a word, a natural-language grammar does well to specify possible head words for those arguments. &amp;quot;Convene&amp;quot; requires an NP object, but some NPs are more semantically or lexically appropriate here than others, and the appropriateness depends largely on the NP's head (e.g., &amp;quot;meeting&amp;quot;). We use the general term bilexical for a grammar that records such facts. A bilexical grammar makes many stipulations about the compatibility of particular pairs of words in particular roles. The acceptability of &amp;quot;Nora convened the &amp;quot; The authors were supported respectively under ARPA Grant N6600194-C-6043 &amp;quot;Human Language Technology&amp;quot; and Ministero dell'Universitk e della Ricerca Scientifica e Tecnologica project &amp;quot;Methodologies and Tools of High Performance Systems for Multimedia Applications.&amp;quot; party&amp;quot; then depends on the grammar writer's assessment of whether parties can be convened.</Paragraph>
    <Paragraph position="3"> Several recent real-world parsers have improved state-of-the-art parsing accuracy by relying on probabilistic or weighted versions of bilexical grammars (Alshawi, 1996; Eisner, 1996; Charniak, 1997; Collins, 1997). The rationale is that soft selectional restrictions play a crucial role in disambiguation, i The chart parsing algorithms used by most of the above authors run in time O(nS), because bilexical grammars are enormous (the part of the grammar relevant to a length-n input has size O(n 2) in practice). Heavy probabilistic pruning is therefore needed to get acceptable runtimes. But in this paper we show that the  grammars where an O(n 3) algorithm was previously known (Eisner, 1997), the grammar constant can be reduced without harming the O(n 3) property.</Paragraph>
    <Paragraph position="4"> Our algorithmic technique throughout is to propose new kinds of subderivations that are not constituents. We use dynamic programming to assemble such subderivations into a full parse.</Paragraph>
  </Section>
class="xml-element"></Paper>