Reliable Weak-to-Strong Monitoring of LLM Agents
Neil Kale , Chen Bo Calvin Zhang , Kevin Zhu , Ankit Aich , Paula Rodriguez , Scale Red Team , Christina Q. Knight , Zifan Wang
- 🏛 Institutions
- Scale AI , Carnegie Mellon University , Massachusetts Institute of Technology
- 📅 Date
- August 26, 2025
- 📑 Publisher
- ICLR 2025
- 💻 Env
- 🔑 Keywords
TLDR
Stress-tests LLM agent monitoring systems for detecting covert misbehavior using a monitor red-teaming (MRT) workflow varying agent/monitor awareness and adversarial evasion strategies, evaluated on SHADE-Arena for tool-calling agents and CUA-SHADE-Arena for computer-use agents.
Related papers (8)
- Genesis: Evolving Attack Strategies for LLM Web Agent Red-TeamingOctober 21, 2025 · ICME 2026
- Human-Guided Harm Recovery for Computer Use AgentsApril 20, 2026 · arXiv
- The Blind Spot of Agent Safety: How Benign User Instructions Expose Critical Vulnerabilities in Computer-Use AgentsApril 12, 2026 · arXiv
- CORA: Conformal Risk-Controlled Agents for Safeguarded Mobile GUI AutomationApril 10, 2026 · arXiv
- Preference Redirection via Attention Concentration: An Attack on Computer Use AgentsApril 9, 2026 · arXiv
- Are GUI Agents Focused Enough? Automated Distraction via Semantic-level UI Element InjectionApril 9, 2026 · arXiv
- Dual-Modality Multi-Stage Adversarial Safety Training: Robustifying Multimodal Web Agents Against Cross-Modal AttacksMarch 4, 2026 · arXiv
- When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use AgentsFebruary 9, 2026 · arXiv