Cordyceps@LT-EDI: Patching Language-Specific Homophobia/Transphobia Classifiers with a Multilingual Understanding

TL;DR

This paper introduces a combined multilingual and language-specific approach to detect homophobia and transphobia in social media, improving accuracy across multiple languages by merging models through interpretable weight interpolation.

Contribution

It presents a novel method for merging multilingual and language-specific classifiers via weight interpolation, enhancing hate speech detection across diverse languages.

Findings

Achieved best results in three of five languages

Attained a 0.997 macro F1-score on Malayalam texts

Demonstrated effectiveness of combined models in real datasets

Abstract

Detecting transphobia, homophobia, and various other forms of hate speech is difficult. Signals can vary depending on factors such as language, culture, geographical region, and the particular online platform. Here, we present a joint multilingual (M-L) and language-specific (L-S) approach to homophobia and transphobic hate speech detection (HSD). M-L models are needed to catch words, phrases, and concepts that are less common or missing in a particular language and subsequently overlooked by L-S models. Nonetheless, L-S models are better situated to understand the cultural and linguistic context of the users who typically write in a particular language. Here we construct a simple and successful way to merge the M-L and L-S approaches through simple weight interpolation in such a way that is interpretable and data-driven. We demonstrate our system on task A of the 'Shared Task on Homophobia/Transphobia Detection in social media comments' dataset for homophobia and transphobic HSD. Our system achieves the best results in three of five languages and achieves a 0.997 macro average F1-score on Malayalam texts.