The exceptional performance of Large Language Models (LLMs) often comes with the unintended propagation of social biases embedded in their training data. While existing benchmarks evaluate overt bias through direct term associations between bias concept terms and demographic terms, LLMs have become increasingly adept at avoiding biased responses, creating an illusion of neutrality. However, biases persist in subtler, contextually hidden forms that traditional benchmarks fail to capture. We introduce the Hidden Bias Benchmark (HBB), a novel dataset designed to assess hidden bias that bias concepts are hidden within naturalistic, subtly framed contexts in real-world scenarios. We analyze six state-of-the-art LLMs, revealing that while models reduce bias in response to overt bias, they continue to reinforce biases in nuanced settings. Data, code, and results are available atthis https URL.
View on arXiv@article{pan2025_2502.19749,
title={ Beneath the Surface: How Large Language Models Reflect Hidden Bias },
author={ Jinhao Pan and Chahat Raj and Ziyu Yao and Ziwei Zhu },
journal={arXiv preprint arXiv:2502.19749},
year={ 2025 }
}