MEDs for PETs: Multilingual Euphemism Disambiguation for Potentially Euphemistic Terms

TL;DR

This paper explores the use of multilingual transformer models to disambiguate potentially euphemistic terms across languages, demonstrating zero-shot learning and cross-lingual transfer benefits in euphemism processing.

Contribution

It introduces a multilingual transformer-based approach for euphemism disambiguation and analyzes cross-lingual transfer and domain-specific effects in euphemism understanding.

Findings

Multilingual models outperform monolingual models significantly.

Zero-shot cross-lingual learning is effective for euphemism disambiguation.

Cross-lingual data within the same domain enhances model performance.

Abstract

This study investigates the computational processing of euphemisms, a universal linguistic phenomenon, across multiple languages. We train a multilingual transformer model (XLM-RoBERTa) to disambiguate potentially euphemistic terms (PETs) in multilingual and cross-lingual settings. In line with current trends, we demonstrate that zero-shot learning across languages takes place. We also show cases where multilingual models perform better on the task compared to monolingual models by a statistically significant margin, indicating that multilingual data presents additional opportunities for models to learn about cross-lingual, computational properties of euphemisms. In a follow-up analysis, we focus on universal euphemistic "categories" such as death and bodily functions among others. We test to see whether cross-lingual data of the same domain is more important than within-language data of other domains to further understand the nature of the cross-lingual transfer.