Watchdogs and Oracles: Runtime Verification Meets Large Language Models for Autonomous Systems

TL;DR

This paper proposes integrating runtime verification with large language models to enhance the safety and reliability of autonomous systems, leveraging their complementary strengths for better assurance.

Contribution

It introduces a novel framework combining runtime verification and LLMs, outlining how their integration can improve specification, anticipation, and uncertainty handling in autonomous systems.

Findings

Highlights the potential of RV and LLMs for dependable autonomy

Discusses challenges and certification issues in integration

Outlines future research directions for safety assurance

Abstract

Assuring the safety and trustworthiness of autonomous systems is particularly difficult when learning-enabled components and open environments are involved. Formal methods provide strong guarantees but depend on complete models and static assumptions. Runtime verification (RV) complements them by monitoring executions at run time and, in its predictive variants, by anticipating potential violations. Large language models (LLMs), meanwhile, excel at translating natural language into formal artefacts and recognising patterns in data, yet they remain error-prone and lack formal guarantees. This vision paper argues for a symbiotic integration of RV and LLMs. RV can serve as a guardrail for LLM-driven autonomy, while LLMs can extend RV by assisting specification capture, supporting anticipatory reasoning, and helping to handle uncertainty. We outline how this mutual reinforcement differs from existing surveys and roadmaps, discuss challenges and certification implications, and identify future research directions towards dependable autonomy.