Relative Bias: A Comparative Framework for Quantifying Bias in LLMs
This work addresses the problem of systematically evaluating bias in LLMs for researchers and practitioners, though it is incremental as it builds on existing comparative analysis approaches.
The paper tackles the challenge of quantifying bias in large language models (LLMs) by proposing the Relative Bias framework, which assesses how an LLM's behavior deviates from others in a target domain using two complementary methods, and finds strong alignment between these methods in case studies.
The growing deployment of large language models (LLMs) has amplified concerns regarding their inherent biases, raising critical questions about their fairness, safety, and societal impact. However, quantifying LLM bias remains a fundamental challenge, complicated by the ambiguity of what "bias" entails. This challenge grows as new models emerge rapidly and gain widespread use, while introducing potential biases that have not been systematically assessed. In this paper, we propose the Relative Bias framework, a method designed to assess how an LLM's behavior deviates from other LLMs within a specified target domain. We introduce two complementary methodologies: (1) Embedding Transformation analysis, which captures relative bias patterns through sentence representations over the embedding space, and (2) LLM-as-a-Judge, which employs a language model to evaluate outputs comparatively. Applying our framework to several case studies on bias and alignment scenarios following by statistical tests for validation, we find strong alignment between the two scoring methods, offering a systematic, scalable, and statistically grounded approach for comparative bias analysis in LLMs.