CIAware-Bench
Explore score history ↗CIAware-Bench measures control intervention awareness in language models through four task domains: essay writing, BigCodeBench, Bash Arena, and SHADE-Arena. Models are tested on distinguishing their own trajectories from those modified by a control protocol, with variations in watermarking, side-task presence, and protocol.
- Released
- 2026-06-09
- Institutions
- Bernstein Center for Computational Neuroscience Tübingen · ELLIS Institute Tübingen · Mila - Quebec Artificial Intelligence Institute
- Evaluates
- General AI, Language & Knowledge, cs.AI
- Openness
- unknown
- Importer
- Claire Radar
- Review status
- unreviewed
CIAware-Bench paper
No reported scores are on record for this benchmark yet.