Mitigating Bias in Queer Representation within Large Language Models: A Collaborative Agent Approach

TL;DR

This paper presents a collaborative agent framework to reduce gender bias in large language models, significantly improving inclusive pronoun usage and promoting fairness in AI-generated content.

Contribution

It introduces a multi-agent pipeline for bias detection and correction in pronoun usage, advancing methods for inclusive language in LLM outputs.

Findings

Achieved a 32.6 percentage point increase in correct inclusive pronoun classification.

Significantly reduced inappropriate gendered pronoun usage in LLM outputs.

Demonstrated the effectiveness of agent-driven frameworks in promoting fairness.

Abstract

Large Language Models (LLMs) often perpetuate biases in pronoun usage, leading to misrepresentation or exclusion of queer individuals. This paper addresses the specific problem of biased pronoun usage in LLM outputs, particularly the inappropriate use of traditionally gendered pronouns ("he," "she") when inclusive language is needed to accurately represent all identities. We introduce a collaborative agent pipeline designed to mitigate these biases by analyzing and optimizing pronoun usage for inclusivity. Our multi-agent framework includes specialized agents for both bias detection and correction. Experimental evaluations using the Tango dataset-a benchmark focused on gender pronoun usage-demonstrate that our approach significantly improves inclusive pronoun classification, achieving a 32.6 percentage point increase over GPT-4o in correctly disagreeing with inappropriate traditionally gendered pronouns $(\chi^2 = 38.57, p < 0.0001)$. These results accentuate the potential of agent-driven frameworks in enhancing fairness and inclusivity in AI-generated content, demonstrating their efficacy in reducing biases and promoting socially responsible AI.