ACL2021

Enhancing Metaphor Detection by Gloss-based Interpretations

Hai Wan, Jinxia Lin, Jianfeng Du, Dawei Shen, Manrong Zhang

Abstract

This paper focuses on utilizing metaphor interpretation to enhance metaphor detection. Considering that existing approaches to metaphor interpretation are limited by ambiguous meanings of the metaphorical substitute words, this paper proposes a novel interpretation mechanism that utilizes glosses to interpret metaphorical words. Since there is no dataset annotated for both metaphor detection and metaphor interpretation, we enhance three datasets TroFi, VUA, and PSUCMC from the field of metaphor detection with gloss annotations. Accordingly, we develop a model for jointly conducting metaphor detection and gloss-based interpretation (named MDGI-Joint for short). Experimental results demonstrate that MDGI-Joint outperforms state-ofthe-art models on all the three enhanced datasets and that gloss-based metaphor interpretation benefits metaphor detection. 1 ⇤ Corresponding author 1 Source code and datasets are available at https:// github.com/sysulic/MDGI. Example : The stroke clouded memories of her youth. Metaphorical word : clouded (Mao, Lin, and Guerin 2018) interpretation : change Meanings of change : make different, remove or replace the coverings of, , become deeper in tone Gloss as interpretation : to make unclear or confused