AOR-Bench
Explore score history ↗AOR-Bench is a benchmark of 3,000 pseudo-harmful audio samples across six categories, designed to evaluate over-refusal in large audio language models. It assesses whether models incorrectly reject benign audio queries.
- Released
- 2026-06-19
- Evaluates
- General AI, Speech & Audio, cs.SD
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
AOR-Bench paper
No reported scores are on record for this benchmark yet.