<?xml version="1.0" standalone="yes"?>
<Paper uid="J94-4005">
  <Title>Training and Scaling Preference Functions for Disambiguation Hiyan Alshawi * AT&amp;T Bell Laboratories</Title>
  <Section position="9" start_page="646" end_page="646" type="ackno">
    <SectionTitle>
Acknowledgments
</SectionTitle>
    <Paragraph position="0"> We would like to thank Manny Rayner for many useful suggestions in carrying out this work, for making the selections necessary to create the database of skeletal trees, and for helping with inspection of experimental results. The paper has also benefited from discussions with Fernando Pereira, Ido Dagan, Michael Collins, Steve Pulman, and Jaan Kaja and from the comments of four anonymous Computational Linguistics referees.</Paragraph>
    <Paragraph position="1"> Most of the work reported here was carried out under a Spoken Language Translation project funded by Telia (formerly Televerket/Swedish Telecom).</Paragraph>
    <Paragraph position="2"> Other parts were done under the CLARE project (JFIT project IED4/1/1165), funded by the UK Department of Trade and Industry, SRI International, the Defence Research Agency, British Telecom, British Petroleum, and British Aerospace.</Paragraph>
  </Section>
class="xml-element"></Paper>