Paper recorded by Signals 4 on 2026-11-02 in cs.CL. Abstract reproduced from arXiv; link to the original below.
Published 2026-11-02 on arXiv · recorded by Signals 4 on 2026-11-03
Category: cs.CL · 自然语言处理 · first seen 2026-11-03
Large language models (LLMs) are increasingly considered for education, recruitment, workplace decision support, and other consequential applications, making pre-deployment social-bias detection essential. Yet major benchmarks are English-centred or translated, focus mainly on negative stereotypes, and may miss linguistic and cultural variation. We investigate gender, racial, and religious bias us