{"id":"arxiv/2605.05134","url":"https://arxiv.org/abs/2605.05134","tags":["cs.LG","math.DS"],"score":0,"title":"Low-Cost Black-Box Detection of LLM Hallucinations via Dynamical System Prediction","domain":"cs.LG","source":"arxiv","authors":["Dan Wilson","Mohamed Akrout"],"reviews":0,"subject":"cs.LG","summary":"Large Language Models (LLMs) frequently generate plausible but non-factual content, a phenomenon known as hallucination. While existing detection methods typically rely on computationally expensive sampling-based consistency checks or external knowledge retrieval, we propose a new method that treats the LLM as a black-box dynamical system. By projecting LLM responses into a high-dimensional manifold via an embedding model, we characterize the resulting vector sequences as observable realizations of the model's latent state-space dynamics. Leveraging Koopman operator theory, we fit the…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-05-06","reactions":{}}