File Information
File: 05-lr/acl_arc_1_sum/cleansed_text/xml_by_section/abstr/06/n06-1058_abstr.xml
Size: 975 bytes
Last Modified: 2025-10-06 13:44:55
<?xml version="1.0" standalone="yes"?> <Paper uid="N06-1058"> <Title>Paraphrasing for Automatic Evaluation</Title> <Section position="1" start_page="0" end_page="0" type="abstr"> <SectionTitle> Abstract </SectionTitle> <Paragraph position="0"> This paper studies the impact of paraphrases on the accuracy of automatic evaluation. Given a reference sentence and a machine-generated sentence, we seek to nd a paraphrase of the reference sentence that is closer in wording to the machine output than the original reference.</Paragraph> <Paragraph position="1"> We apply our paraphrasing method in the context of machine translation evaluation.</Paragraph> <Paragraph position="2"> Our experiments show that the use of a paraphrased synthetic reference re nes the accuracy of automatic evaluation. We also found a strong connection between the quality of automatic paraphrases as judged by humans and their contribution to automatic evaluation.</Paragraph> </Section> class="xml-element"></Paper>