ART-SafeBench
Explore score history ↗ART-SafeBench evaluates attacks against multimodal agentic RAG systems across four attack surfaces, using attack success rate and query cost as comparable outcomes.
- Released
- 2026-06-25
- Evaluates
- Language & Knowledge, Robotics & Embodied AI, cs.CR
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- ai-reviewed
ART-SafeBench paper and code
No reported scores are on record for this benchmark yet.