SOURCE-LINKED INTELLIGENCE

When Do Supervised UQ Ensembles Improve LLM Hallucination Detection? A Robustness Study

arXiv · AI, language, vision and robotics · article · Aug 25, 2026 · UTC

Uncertainty quantification (UQ) methods are widely used for hallucination detection in large language models (LLMs) in closed-book settings where ground-truth evidence is unavailable at inference time. Prior work has proposed combining UQ signals via learned ensembles, but empirical investigations into the robustness of these ensembles are limited. We study a supervised ensembling framework that trains a classifier over heterogeneous UQ-based scorer outputs on a small, domain-specific dataset of labeled LLM responses, then applies it to out-of-sample hallucination classification without retrie

Read original source ↗ Open in workspace

recordType

paper

region

Global