ContextBench
Explore score history ↗Evaluates text-to-image models on 92 roles and 1,656 controlled prompts across context-free, related, and unrelated conditions, measuring bias persistence in visual attributes through metrics like Bias Intensity and Context Consistency Score.
- Released
- 2026-08-30
- Evaluates
- General AI, Safety & Trustworthiness, cs.CV
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
ContextBench paper, code and dataset
No reported scores are on record for this benchmark yet.