Grounded noun/verb-phrases to images for RTE

韓 丹; マルティネス パスクアル

doi:10.11517/pjsai.JSAI2017.0_4Q14in2

Proceedings of the Annual Conference of JSAI

Online ISSN : 2758-7347

31st (2017)

Session ID : 4Q1-4in2

DOI https://doi.org/10.11517/pjsai.JSAI2017.0_4Q14in2

Conference information

Host: The Japanese Society for Artificial Intelligence

Name : The 31st Annual Conference of the Japanese Society for Artificial Intelligence, 2017

Number : 31

Location : [in Japanese]

Date : May 23, 2017 - May 26, 2017

Grounded noun/verb-phrases to images for RTE

*[in Japanese], [in Japanese]

Author information

CONFERENCE PROCEEDINGS FREE ACCESS

Details

Abstract

We propose an algorithm to ground noun phrases and verb phrases to images, and deduce the relations between phrases from examining the similarity of two sets of images according to a simple similarity measure and a threshold. We evaluate our method in a textual entailment task.

Content from these authors

Favorites & Alerts

Corresponding author

Conference information

Register with J-STAGE for free!