VIBE: Voice-Induced open-ended Bias Evaluation for Large Audio-Language Models via Real-World Speech

ArXi:2604.17248v1 Announce Type: cross Large Audio-Language Models (LALMs) are increasingly integrated into daily applications, yet their generative biases remain underexplored. Existing speech fairness benchmarks rely on synthetic speech and Multiple-Choice Questions (MCQs), both offering a fragmented view of fairness. We propose VIBE, a framework that evaluates generative bias through open-ended tasks such as personalized recommendations, using real-world human recordings.