Please use this identifier to cite or link to this item:
Title: SWIM: A simple word interaction model for implicit discourse relation recognition
Authors: Lei, W 
Wang, X
Liu, M
Ilievski, I 
He, X 
Kan, MY 
Issue Date: Aug-2017
Publisher: International Joint Conferences on Artificial Intelligence Organization
Citation: Lei, W, Wang, X, Liu, M, Ilievski, I, He, X, Kan, MY (2017-08). SWIM: A simple word interaction model for implicit discourse relation recognition. Twenty-Sixth International Joint Conference on Artificial Intelligence 0 : 4026-4032. ScholarBank@NUS Repository.
Abstract: Capturing the semantic interaction of pairs of words across arguments and proper argument representation are both crucial issues in implicit discourse relation recognition. The current state-of-the-art represents arguments as distributional vectors that are computed via bi-directional Long Short-Term Memory networks (BiLSTMs), known to have significant model complexity. In contrast, we demonstrate that word-weighted averaging can encode argument representation which can be incorporated with word pair information efficiently. By saving an order of magnitude in parameters and eschewing the recurrent structure, our proposed model achieves equivalent performance, but trains seven times faster.
Source Title: Twenty-Sixth International Joint Conference on Artificial Intelligence
ISBN: 9780999241103
ISSN: 10450823
DOI: 10.24963/ijcai.2017/562
Appears in Collections:Staff Publications

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
swim-simple-word.pdf1.12 MBAdobe PDF



Google ScholarTM



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.