UMass PCL at SemEval-2022 Task 4: Pre-trained Language Model Ensembles for Detecting Patronizing and Condescending Language

TL;DR

This paper presents an ensemble approach using pre-trained language models, data augmentation, and threshold optimization to detect patronizing and condescending language, achieving competitive results in a SemEval challenge.

Contribution

It introduces a novel ensemble system leveraging pre-trained models and data augmentation for detecting subtle PCL in media, addressing data scarcity challenges.

Findings

Achieved 55.47% F1 score on binary classification

Attained 36.25% macro F1 on multi-label detection

Demonstrated reliable detection of subtle patronizing language

Abstract

Patronizing and condescending language (PCL) is everywhere, but rarely is the focus on its use by media towards vulnerable communities. Accurately detecting PCL of this form is a difficult task due to limited labeled data and how subtle it can be. In this paper, we describe our system for detecting such language which was submitted to SemEval 2022 Task 4: Patronizing and Condescending Language Detection. Our approach uses an ensemble of pre-trained language models, data augmentation, and optimizing the threshold for detection. Experimental results on the evaluation dataset released by the competition hosts show that our work is reliably able to detect PCL, achieving an F1 score of 55.47% on the binary classification task and a macro F1 score of 36.25% on the fine-grained, multi-label detection task.