Benchmark
SteerBench
What is SteerBench?
SteerBench evaluates full-duplex spoken steerability using 390 spoken prompts and 1,067 human-authored binary audio and text rubrics covering tone, persona…
- Released
- 2026-09-11
- Evaluates
- General AI, Speech & Audio, cs.AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-name-audit-deferred
- Reported scores
- 0
Source provenance
- Original evidence https://arxiv.org/abs/2609.12623
SteerBench paper
No reported scores are on record for this benchmark yet.