AIVV: Neuro-Symbolic LLM Agent-Integrated Verification and Validation for Trustworthy Autonomous Systems

2026-04-02

AI Summary

This work addresses the challenge that traditional deep learning approaches struggle to distinguish genuine faults from noise or transient disturbances in anomaly detection, leading to verification and validation (V&V) processes that are heavily reliant on manual effort and lack scalability. To overcome this limitation, the paper introduces the AIVV framework, which pioneers the integration of neuro-symbolic reasoning with a collaborative multi-agent architecture based on large language models (LLMs) into autonomous system V&V workflows. Specifically, a committee of role-specialized LLM agents performs semantic validation of anomalies and automatically generates actionable V&V artifacts directly from natural language requirements. Evaluated in a temporal simulation environment for unmanned underwater vehicles, the approach successfully digitizes manual V&V procedures and substantially outperforms rule-based fault classification methods, establishing a scalable, high-fidelity paradigm for LLM-mediated supervision in time-series data domains.

Abstract

Deep learning models excel at detecting anomaly patterns in normal data. However, they do not provide a direct solution for anomaly classification and scalability across diverse control systems, frequently failing to distinguish genuine faults from nuisance faults caused by noise or the control system's large transient response. Consequently, because algorithmic fault validation remains unscalable, full Verification and Validation (V\&V) operations are still managed by Human-in-the-Loop (HITL) analysis, resulting in an unsustainable manual workload. To automate this essential oversight, we propose Agent-Integrated Verification and Validation (AIVV), a hybrid framework that deploys Large Language Models (LLMs) as a deliberative outer loop. Because rigorous system verification strictly depends on accurate validation, AIVV escalates mathematically flagged anomalies to a role-specialized LLM council. The council agents perform collaborative validation by semantically validating nuisance and true failures based on natural-language (NL) requirements to secure a high-fidelity system-verification baseline. Building on this foundation, the council then performs system verification by assessing post-fault responses against NL operational tolerances, ultimately generating actionable V\&V artifacts, such as gain-tuning proposals. Experiments on a time-series simulator for Unmanned Underwater Vehicles (UUVs) demonstrate that AIVV successfully digitizes the HITL V\&V process, overcoming the limitations of rule-based fault classification and offering a scalable blueprint for LLM-mediated oversight in time-series data domains.

Problem

• Verification and Validation
• anomaly classification
• Human-in-the-Loop
• fault validation
• scalability

Innovation

• Neuro-Symbolic
• Large Language Models
• Verification and Validation
• Autonomous Systems
• Anomaly Classification