Bias Benchmark for QA (BBQ)

The 'Bias Benchmark for QA (BBQ)' is a dataset designed to evaluate biases in question-answering systems by assessing their responses to prompts that include explicitly stated protected attributes such as religion, race, and gender.

Papers using Bias Benchmark for QA (BBQ) (2)

"The Dentist is an involved parent, the bartender is not": Revealing Implicit Biases in QA with Implicit BBQ2025Obscured but Not Erased: Evaluating Nationality Bias in LLMs via Name-Based Bias Benchmarks2025