ReactHuman Benchmark
Explore score history ↗Evaluates human-like reactive decision-making in multimodal LLMs by simulating 1,000+ reproducible household hazard scenes with physics-grounded ground truth across 17 event families.
- Released
- 2026-09-09
- Evaluates
- Multimodal, Robotics & Embodied AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
ReactHuman paper and dataset
No reported scores are on record for this benchmark yet.