Transformer Ensembles for Sexism Detection

TL;DR

This paper explores ensemble methods using Transformer models for sexism detection, achieving competitive accuracy and F1 scores on the EXIST2021 dataset.

Contribution

It introduces an ensemble approach with Transformer models trained on diverse backgrounds for improved sexism detection.

Findings

Accuracy of 0.767 on binary classification

F1 score of 0.766 on binary classification

Accuracy of 0.623 on multi-class classification

Abstract

This document presents in detail the work done for the sexism detection task at EXIST2021 workshop. Our methodology is built on ensembles of Transformer-based models which are trained on different background and corpora and fine-tuned on the provided dataset from the EXIST2021 workshop. We report accuracy of 0.767 for the binary classification task (task1), and f1 score 0.766, and for the multi-class task (task2) accuracy 0.623 and f1-score 0.535.