CURA: Certified Runtime Alarms for Computer-Use Agents
arXiv cs.AIen
arXiv cs.AI
AI Global WirearXiv:2608.27808v1 Announce Type: new Abstract: Self-report is the cheapest oversight channel a deployer has, and on capable computer-use agents (CUAs) it fails precisely where oversight matters. On 361 OSWorld tasks our pipeline, a read-only feasibility gate, a planner, and a GUI executor, reaches a mean task score of 82.9, above the 72.4 human reference, yet 64 of its 71 failures (90%) end with a success claim, 61 acknowledging no blocker, and the explicit failure affordance is never used in roughly 9,100 calls. We introduce CURA (Certified Runtime Alarms for Computer-Use Agents), an external monitor that reads only harness-visible telemetry, with no model internals, extra LLM calls, or pr
This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.
Read the full story at arXiv cs.AI- Forskning
- Agenter
Related AI news
- Big Tech reported Q2 "other income" rose significantly to $160B+, driven by investments in AI companies, raising concerns of paper gains overstating the AI boom (Financial Times)Techmeme · August 31, 2026
- Life’s a game, and AI agents are Animoca’s new playTech in Asia · August 31, 2026
- Effectiveness of IoT and Deep Learning for Detection and Severity Assessment of Postelectrotermes militaris in Tea PlantationsarXiv cs.AI · August 31, 2026
- Thinking Costs Tokens: When More Structure is Worth the PricearXiv cs.AI · August 31, 2026
- Nemotron 3.5 Content Safety Moderator: A Compact Multimodal, Multilingual, and Reasoning Enabled Content Safety ModeratorarXiv cs.AI · August 31, 2026
- Probing Perceptual Priors of MLLMs via Gibbs Sampling with Interpretable Generative ControlsarXiv cs.AI · August 31, 2026