Containing Behavioral Cascades from Manipulated Claims in LLM-Powered Multi-Robot Systems

2026-09-24

AI Summary

This study addresses the problem of false claims in LLM-driven multi-robot systems triggering behavioral cascades that lead to costly replanning and task failures. To mitigate this, we propose a proactive verification framework that formulates post-hoc verification as a team-level planning problem rather than a binary trust decision. The framework generates "verify-adapt-maintain" plans that contain the downstream propagation of semantic manipulation through local checks and bounded adaptations. By integrating structured verification modules with adaptive path planning, the proposed method significantly reduces overall system cost and makespan while improving task coverage in multi-robot transportation scenarios. Ultimately, this approach effectively constrains the spread of physical consequences and suppresses cascading failures across the robotic team.

Abstract

Large language model (LLM)-powered multi-robot systems are vulnerable to semantic manipulation: an accepted false world-state claim can trigger a fleet-wide behavioral cascade, causing unnecessary replanning, increased path costs, congestion, or apparent mission infeasibility. Conditioning on a successful manipulation, we propose an active verification framework that contains its downstream effects before they propagate across the fleet. A dedicated verification module generates a structured Verify-Adapt-Hold plan: selected robots inspect consequential regions, a limited subset provisionally adapts when necessary, and the remaining robots retain their trusted plans. We evaluate the framework in a multi-robot transportation environment using injected false obstacle claims across different impacts and team sizes. Evaluation measures cascade containment, Sum-of-Costs, makespan, and coverage ratio. Results show that treating post-compromise verification as a team-level planning problem, rather than a binary trust decision, effectively limits the cascading physical consequences of semantic manipulation.

Problem

• Large Language Models
• Multi-Robot Systems
• Semantic Manipulation
• Behavioral Cascades
• False Claims

Innovation

• Behavioral Cascades
• Semantic Manipulation
• Active Verification Framework
• Multi-Robot Systems
• Large Language Models