Benchmark RadarAI BENCHMARK PROFILE
ADeptS-Bench
Computer Use Agents (CUAs) are increasingly deployed to navigate mobile and desktop applications on behalf of users, yet no benchmark comprehensively evaluates whether they can safely interact with visual interfaces while handling ambiguou…
- Released
- 2026-08-25
- Readiness
- Paper only
- Primary field
- Industrial & Engineering
Motivation
Computer Use Agents (CUAs) are increasingly deployed to navigate mobile and desktop applications on behalf of users, yet no benchmark comprehensively evaluates whether they can safely interact with visual interfaces while handling ambiguous instructions.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.