EigenTrack: Spectral Activation Feature Tracking for Hallucination and Out-of-Distribution Detection in LLMs and VLMs
Large language models (LLMs) offer broad utility but remain prone to hallucination and out-of-distribution (OOD) errors. We propose EigenTrack, an interpretable real-time detector that uses the spectral geometry of hidden activations, a compact global signature of model dynamics. By streaming covariance-spectrum statistics such as entropy, eigenvalue gaps, and KL divergence from random baselines into a lightweight recurrent classifier, EigenTrack tracks temporal shifts in representation structure that signal hallucination and OOD drift before surface errors appear. Unlike black- and grey-box methods, it needs only a single forward pass without resampling. Unlike existing white-box detectors, it preserves temporal context, aggregates global signals, and offers interpretable accuracy-latency trade-offs.
Code (0)
등록된 구현이 없습니다.
Tasks
Out-of-Distribution DetectionSimilar Papers 제목 키워드 기반
Spectral Geometry for Deep Learning: Compression and Hallucination Detection via Random Matrix Theory
Large language models and deep neural networks achieve strong performance but suffer from reliability issues and high computational cost. This thesis proposes a unified framework based on spectral geometry and random mat…
Knowledge DistillationStructure and Redundancy in Large Language Models: A Spectral Study via Random Matrix Theory
This thesis addresses two persistent and closely related challenges in modern deep learning, reliability and efficiency, through a unified framework grounded in Spectral Geometry and Random Matrix Theory (RMT). As deep n…
Knowledge DistillationFrom Dispersion to Attraction: Spectral Dynamics of Hallucination Across Whisper Model Scales
Hallucinations in large ASR models present a critical safety risk. In this work, we propose the \textit{Spectral Sensitivity Theorem}, which predicts a phase transition in deep networks from a dispersive regime (signal d…
Spectral Guardrails for Agents in the Wild: Detecting Tool Use Hallucinations via Attention Topology
Deploying autonomous agents in the wild requires reliable safeguards against tool use failures. We propose a training free guardrail based on spectral analysis of attention topology that complements supervised approaches…
LLM Hallucination Detection: A Fast Fourier Transform Method Based on Hidden Layer Temporal Signals
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 fundam…