<?xml version="1.0" standalone="yes"?>
<Paper uid="E06-2015">
  <Title>Semantic Role Labeling for Coreference Resolution</Title>
  <Section position="2" start_page="0" end_page="0" type="intro">
    <SectionTitle>
1 Introduction
</SectionTitle>
    <Paragraph position="0"> The last years have seen a boost of work devoted to the development of machine learning based coreference resolution systems (Soon et al., 2001; Ng &amp; Cardie, 2002; Kehler et al., 2004, inter alia).</Paragraph>
    <Paragraph position="1"> Similarly, many researchers have explored techniques for robust, broad coverage semantic parsing in terms of semantic role labeling (Gildea &amp; Jurafsky, 2002; Carreras &amp; M`arquez, 2005, SRL henceforth).</Paragraph>
    <Paragraph position="2"> This paper explores whether coreference resolution can benefit from SRL, more specifically, which phenomena are affected by such information. The motivation comes from the fact that current coreference resolution systems are mostly relying on rather shallow features, such as the distance between the coreferent expressions, string matching, and linguistic form. On the other hand, the literature emphasizes since the very beginning the relevance of world knowledge and inference (Charniak, 1973). As an example, consider a sentence from the Automatic Content Extraction  (ACE) 2003 data.</Paragraph>
    <Paragraph position="3"> (1) A state commission of inquiry into the sinking of the  Kursk will convene in Moscow on Wednesday, the Interfax news agency reported. It said that the diving operation will be completed by the end of next week. It seems that in this example, knowing that the Interfax news agency is the AGENT of the report predicate, and It being the AGENT of say, could trigger the (semantic parallelism based) inference required to correctly link the two expressions, in contrast to anchoring the pronoun to Moscow.</Paragraph>
    <Paragraph position="4"> SRL provides the semantic relationships that constituents have with predicates, thus allowing us to include document-level event descriptive information into the relations holding between referring expressions (REs). This layer of semantic context abstracts from the specific lexical expressions used, and therefore represents a higher level of abstraction than predicate argument statistics (Kehler et al., 2004) and Latent Semantic Analysisusedasamodelofworldknowledge(Klebanov null &amp; Wiemer-Hastings, 2002). In this respect, the present work is closer in spirit to Ji et al. (2005), who explore the employment of the ACE 2004 relation ontology as a semantic filter.</Paragraph>
  </Section>
class="xml-element"></Paper>