LLM Hallucination Detection: A Fast Fourier Transform Method Based on Hidden Layer Temporal Signals
This addresses the critical barrier of hallucination for deploying LLMs in reliability-sensitive applications, representing a new paradigm rather than an incremental improvement.
The paper tackles the problem of hallucination detection in large language models by proposing HSAD, a framework that models temporal dynamics of hidden representations using Fast Fourier Transform, achieving over 10 percentage points improvement on benchmarks like TruthfulQA compared to prior state-of-the-art methods.
Hallucination remains a critical barrier for deploying large language models (LLMs) in reliability-sensitive applications. Existing detection methods largely fall into two categories: factuality checking, which is fundamentally constrained by external knowledge coverage, and static hidden-state analysis, that fails to capture deviations in reasoning dynamics. As a result, their effectiveness and robustness remain limited. We propose HSAD (Hidden Signal Analysis-based Detection), a novel hallucination detection framework that models the temporal dynamics of hidden representations during autoregressive generation. HSAD constructs hidden-layer signals by sampling activations across layers, applies Fast Fourier Transform (FFT) to obtain frequency-domain representations, and extracts the strongest non-DC frequency component as spectral features. Furthermore, by leveraging the autoregressive nature of LLMs, HSAD identifies optimal observation points for effective and reliable detection. Across multiple benchmarks, including TruthfulQA, HSAD achieves over 10 percentage points improvement compared to prior state-of-the-art methods. By integrating reasoning-process modeling with frequency-domain analysis, HSAD establishes a new paradigm for robust hallucination detection in LLMs.