Abstract

The assumption that natural language expressions have a single, discrete and clearly identifiable meaning in a given context, successfully challenged in lexical semantics by the rise of distributional models, nevertheless still underlies much work in computational linguistics, including work based on distributed representations. In this talk I will first of all present the evidence that convinced us that the assumption that a single interpretation can always be assigned to anaphoric expression is no more than a convenient idealization. I will then discuss recent work on the DALI project that aims to develop a new model of interpretation that abandons this assumption for the case of anaphoric interpretaton / coreference. I will present the recently released Phrase Detectives 2.1 corpus, containing around 2 million crowdsourced judgments for more than 100,000 markables, an average of 20 judgments per markable; the Mention Pair Annotation (MPA) Bayesian inference model developed to aggregate these judgments; and the results of a preliminary analysis of disagreements in the corpus suggesting that between 10% and 30% of markables in the corpus appear to be genuinely ambiguous.

Joint work with Jon Chamberlain, Silviu Paun, Alexandra Uma, Juntao Yu, Richard Bartle and Udo Kruschwitz