Social Bias Benchmark for Generation: A Comparison of Generation and QA-Based Evaluations

Warning: This paper contains examples of stereotypes and biases. Measuring social bias in large language models (LLMs) is crucial, but existing bias evaluation methods struggle to assess bias in longform generation. We propose a Bias Benchmark for Generation (BBG), an adaptation of the Bias Benchmark for QA (BBQ), designed to evaluate social bias in long-form generation by having LLMs generate continuations of story prompts. Building our benchmark in English and Korean, we measure the probability of neutral and biased generations across ten LLMs. We also compare our long-form story generation evaluation results with multiplechoice BBQ evaluation, showing that the two approaches produce inconsistent results.