Detecting Euphemisms with Literal Descriptions and Visual Imagery

TL;DR

This paper presents a two-stage system for euphemism detection that combines literal descriptions and visual imagery to improve accuracy, achieving second place in a shared task with an F1 score of 87.2%.

Contribution

The paper introduces a novel two-stage approach that incorporates literal descriptions and visual imagery to enhance euphemism detection performance.

Findings

Literal descriptions significantly improve detection accuracy.

Visual supervision provides additional performance gains.

System achieved second place with an F1 score of 87.2%.

Abstract

This paper describes our two-stage system for the Euphemism Detection shared task hosted by the 3rd Workshop on Figurative Language Processing in conjunction with EMNLP 2022. Euphemisms tone down expressions about sensitive or unpleasant issues like addiction and death. The ambiguous nature of euphemistic words or expressions makes it challenging to detect their actual meaning within a context. In the first stage, we seek to mitigate this ambiguity by incorporating literal descriptions into input text prompts to our baseline model. It turns out that this kind of direct supervision yields remarkable performance improvement. In the second stage, we integrate visual supervision into our system using visual imageries, two sets of images generated by a text-to-image model by taking terms and descriptions as input. Our experiments demonstrate that visual supervision also gives a statistically significant performance boost. Our system achieved the second place with an F1 score of 87.2%, only about 0.9% worse than the best submission.