SafeRelBench
Explore score history ↗SafeRelBench evaluates VLM-driven embodied agents on household tasks with process-level safety constraints, focusing on spatial relations such as support, containment, and proximity. It includes 507 executable samples (248 spatial-relation, 259 control) and measures whether agents satisfy safety conditions before risk-prone actions, alongside task success.
- Released
- 2026-07-16
- Institutions
- Beijing University of Posts and Telecommunications · Beijing Institute for General Artificial Intelligence
- Evaluates
- Robotics & Embodied AI, Safety
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
SafeRelBench paper
No reported scores are on record for this benchmark yet.