SENTINEL: A Multi-Level Formal Framework for Safety Evaluation of LLM-based Embodied Agents
By: Simon Sinong Zhan , Yao Liu , Philip Wang and more
Potential Business Impact:
Keeps robots from doing dangerous things.
We present Sentinel, the first framework for formally evaluating the physical safety of Large Language Model(LLM-based) embodied agents across the semantic, plan, and trajectory levels. Unlike prior methods that rely on heuristic rules or subjective LLM judgments, Sentinel grounds practical safety requirements in formal temporal logic (TL) semantics that can precisely specify state invariants, temporal dependencies, and timing constraints. It then employs a multi-level verification pipeline where (i) at the semantic level, intuitive natural language safety requirements are formalized into TL formulas and the LLM agent's understanding of these requirements is probed for alignment with the TL formulas; (ii) at the plan level, high-level action plans and subgoals generated by the LLM agent are verified against the TL formulas to detect unsafe plans before execution; and (iii) at the trajectory level, multiple execution trajectories are merged into a computation tree and efficiently verified against physically-detailed TL specifications for a final safety check. We apply Sentinel in VirtualHome and ALFRED, and formally evaluate multiple LLM-based embodied agents against diverse safety requirements. Our experiments show that by grounding physical safety in temporal logic and applying verification methods across multiple levels, Sentinel provides a rigorous foundation for systematically evaluating LLM-based embodied agents in physical environments, exposing safety violations overlooked by previous methods and offering insights into their failure modes.
Similar Papers
Sentinel Agents for Secure and Trustworthy Agentic AI in Multi-Agent Systems
Artificial Intelligence
Protects smart systems from bad actors.
AgentSentinel: An End-to-End and Real-Time Security Defense Framework for Computer-Use Agents
Cryptography and Security
Stops smart computer helpers from doing bad things.
OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflows
Artificial Intelligence
Finds dangerous actions by phone apps.