Siamese Convolutional Neural Networks for Recognizing Partial Entailment

Warning

This publication doesn't include Faculty of Economics and Administration. It includes Faculty of Informatics. Official publication website can be found on muni.cz.
Authors

VÍTA Martin

Year of publication 2018
Type Article in Proceedings
Conference Siamese Convolutional Neural Networks for Recognizing Partial Entailment
MU Faculty or unit

Faculty of Informatics

Citation
Web Full paper
Keywords Partial Textual Entailment; Convolutional Neural Networks; Siamese Architectures
Description Recognizing textual entailment (RTE), i. e., a decision problem whether a sentence (called hypothesis) can be inferred from a given text, became a well established and widely studied task. As a consequence of the traditional binary (or ternary) class formulation, it is not possible to express the fact that a fragment of the hypothesis is entailed by the text, even though the “whole” entailment of the hypothesis from the text does not hold. The notions of partial textual entailment – and faceted entailment in particular – address this problem. In this paper, we introduce a siamese CNN architecture with a static attention mechanism together with a sentence compression and provide an evaluation over modified SemEval 2013 Task 8 dataset.
Related projects:

You are running an old browser version. We recommend updating your browser to its latest version.