Semantic Invariance in Agentic AI

ArXi:2603.13173v1 Announce Type: new Large Language Models (LLMs) increasingly serve as autonomous reasoning agents in decision, scientific problem-solving, and multi-agent coordination systems. However, deploying LLM agents in consequential applications requires assurance that their reasoning remains stable under semantically equivalent input variations, a property we term semantic invariance. Standard benchmark evaluations, which assess accuracy on fixed, canonical problem formulations, fail to capture this critical reliability dimension.