Runtime Authorization of Self-Generated Subgoals in Long-Horizon Tool-Using AI Agents
cs.AI, cs.CR
Submitted: 2026-10-04
Updated: 2026-10-04
Comments: 38 pages, 3 figures, 8 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Agent Behavioral Contracts: Formal Specification and Runtime Enforcement for Reliable Autonomous AI Agents
- SafetyDrift: Predicting When AI Agents Cross the Line Before They Actually Do
- Enforcing Temporal Constraints for LLM Agents
- Safe Multi-Agent Behavior Must Be Maintained, Not Merely Asserted: Constraint Drift in LLM-Based Multi-Agent Systems
- VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills
- ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis
- CaMeLoT: CaMeL orchestrated with Temporal logic for static verification and liveness
- HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark
- Structural Enforcement of Goal Integrity in AI Agents via Separation-of-Powers Architecture
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems