DR.GAP: Mitigating Bias in Large Language Models using Gender-Aware Prompting with Demonstration and Reasoning

TL;DR

DR.GAP is a novel, automated prompting method that reduces gender bias in large language models without sacrificing performance, applicable across various models and tasks.

Contribution

It introduces a model-agnostic, prompt-based approach that mitigates gender bias by selecting bias-revealing examples and generating structured reasoning.

Findings

Effective bias reduction across multiple LLMs and tasks

Maintains model utility while reducing bias

Generalizes to vision-language models

Abstract

Large Language Models (LLMs) exhibit strong natural language processing capabilities but also inherit and amplify societal biases, including gender bias, raising fairness concerns. Existing debiasing methods face significant limitations: parameter tuning requires access to model weights, prompt-based approaches often degrade model utility, and optimization-based techniques lack generalizability. To address these challenges, we propose DR.GAP (Demonstration and Reasoning for Gender-Aware Prompting), an automated and model-agnostic approach that mitigates gender bias while preserving model performance. DR.GAP selects bias-revealing examples and generates structured reasoning to guide models toward more impartial responses. Extensive experiments on coreference resolution and QA tasks across multiple LLMs (GPT-3.5, Llama3, and Llama2-Alpaca) demonstrate its effectiveness, generalization ability, and robustness. DR.GAP can generalize to vision-language models (VLMs), achieving significant bias reduction.