Leveraging Large Language Models and Topic Modeling for Toxicity Classification

TL;DR

This paper investigates how fine-tuning large language models with topic modeling can improve toxicity classification, revealing limitations of current models and emphasizing the influence of annotator bias on model performance.

Contribution

It introduces a combined approach of fine-tuning BERTweet and HateBERT with topic modeling to enhance toxicity detection accuracy.

Findings

Fine-tuning models on specific topics improves F1 scores.

State-of-the-art models still struggle with toxicity detection accuracy.

Annotator bias impacts model training and outcomes.

Abstract

Content moderation and toxicity classification represent critical tasks with significant social implications. However, studies have shown that major classification models exhibit tendencies to magnify or reduce biases and potentially overlook or disadvantage certain marginalized groups within their classification processes. Researchers suggest that the positionality of annotators influences the gold standard labels in which the models learned from propagate annotators' bias. To further investigate the impact of annotator positionality, we delve into fine-tuning BERTweet and HateBERT on the dataset while using topic-modeling strategies for content moderation. The results indicate that fine-tuning the models on specific topics results in a notable improvement in the F1 score of the models when compared to the predictions generated by other prominent classification models such as GPT-4, PerspectiveAPI, and RewireAPI. These findings further reveal that the state-of-the-art large language models exhibit significant limitations in accurately detecting and interpreting text toxicity contrasted with earlier methodologies. Code is available at https://github.com/aheldis/Toxicity-Classification.git.