We propose a neural machine translation architecture that models the surrounding text in addition to the source sentence. These models lead to better performance, both in terms of general translation quality and pronoun prediction, when trained on small corpora, although this improvement largely disappears when trained with a larger corpus. We also discover that attention-based neural machine t...

Linguists draw a distinction between two types of interrogatives: discourse linked (d-linked) phrases such as which man, which implies the existence of a set of contextually determined entities (men) from which the speaker is asking for a choice, and non-d-linked interrogatives such as who, which carry no such implication. Two questionnaires and an on-line reading study showed that readers pref...

We investigated the processing of pronouns in Strong and Weak Crossover constructions as a means of probing the extent to which the incremental parser can use syntactic information to guide antecedent retrieval. In Experiment 1 we show that the parser accesses a displaced wh-phrase as an antecedent for a pronoun when no grammatical constraints prohibit binding, but the parser ignores the same w...

This paper examines the time course of the use of implicit verb causality and gender agreement information in determining the antecedent of a pronominal anaphor. A crossmodal probe paradigm measured the activation of candidate antecedents at four time points during comprehension. Sentences contained two participants, an implicit causality verb, and a subsequent “because”-clause (e.g.,Gary amaze...

In this paper, we employ the centering theory in pronoun resolution from the semantic perspective. First, diverse semantic role features with regard to different predicates in a sentence are explored. Moreover, given a pronominal anaphor, its relative ranking among all the pronouns in a sentence, according to relevant semantic role information and its surface position, is incorporated. In parti...

Most existing approaches for zero pronoun resolution are supervised approaches, where annotated data are released by shared task organizers. Therefore, the lack of annotated data becomes a major obstacle in zero pronoun resolution task. The existing approaches mainly face the challenge of costing manpower on labeling the extended data for better training performance and domain adaption. To alle...

This paper proposes a context-sensitive convolution tree kernel for pronoun resolution. It resolves two critical problems in previous researches in two ways. First, given a parse tree and a pair of an anaphor and an antecedent candidate, it implements a dynamic-expansion scheme to automatically determine a proper tree span for pronoun resolution by taking predicateand antecedent competitor-rela...

This paper addresses the translation divergencies of pronouns from English to French, specifically it and they, which have several gendered and non-gendered possible translations into French. Instead of using anaphora resolution, which is error-prone, we build a target language model that estimates the probabilities of a tuple of consecutive nouns followed by a pronoun. We bring evidence for th...