Stereotype Detection as a Catalyst for Enhanced Bias Detection: A Multi-Task Learning Approach

TL;DR

This paper proposes a multi-task learning approach for bias and stereotype detection in language models, demonstrating that joint training improves bias detection performance and leveraging stereotype information enhances fairness.

Contribution

It introduces StereoBias, a new dataset for bias and stereotype detection, and shows that joint learning of these tasks improves bias detection accuracy across different model architectures.

Findings

Joint training significantly improves bias detection performance.

Encoder-only models perform well, but decoder-only models are competitive.

Leveraging stereotype information enhances fairness in AI systems.

Abstract

Bias and stereotypes in language models can cause harm, especially in sensitive areas like content moderation and decision-making. This paper addresses bias and stereotype detection by exploring how jointly learning these tasks enhances model performance. We introduce StereoBias, a unique dataset labeled for bias and stereotype detection across five categories: religion, gender, socio-economic status, race, profession, and others, enabling a deeper study of their relationship. Our experiments compare encoder-only models and fine-tuned decoder-only models using QLoRA. While encoder-only models perform well, decoder-only models also show competitive results. Crucially, joint training on bias and stereotype detection significantly improves bias detection compared to training them separately. Additional experiments with sentiment analysis confirm that the improvements stem from the connection between bias and stereotypes, not multi-task learning alone. These findings highlight the value of leveraging stereotype information to build fairer and more effective AI systems.