A deep network with visual text composition behavior

From National Research Council Canada

DOIResolve DOI: https://doi.org/10.18653/v1/P17-2059
AuthorSearch for: 1
Name affiliation
  1. National Research Council of Canada. Information and Communication Technologies
FormatText, Article
Proceedings titleProceedings of the 55th Annual Meeting of the Association for Computational Linguistics. Volume 2: Short Papers
ConferenceThe 55th Annual Meeting of the Association for Computational Linguistics, 30 July - August 4, 2017, Vancouver, BC., Canada
Pages372–377
Abstract
While natural languages are composi- tional, how state-of-the-art neural models achieve compositionality is still unclear. We propose a deep network, which not only achieves competitive accuracy for text classification, but also exhibits compositional behavior. That is, while creating hierarchical representations of a piece of text, such as a sentence, the lower layers of the network distribute their layer-specific attention weights to individual words. In contrast, the higher layers compose meaningful phrases and clauses, whose lengths increase as the networks get deeper until fully composing the sentence.
Publication date
PublisherAssociation for Computational Linguistics
LanguageEnglish
Peer reviewedYes
NPARC number23002276
Export citationExport as RIS
Report a correctionReport a correction
Record identifiera6bdaa46-5f05-4281-8ccc-bf160e7a5aff
Record created2017-09-28
Record modified2020-03-16
Report a problem on this page
Date modified: