KEA Explain: Explanations of Hallucinations using Graph Kernel Analysis

TL;DR

This paper introduces KEA Explain, a neurosymbolic framework that detects and explains hallucinations in large language models by comparing knowledge graphs from model outputs with ground truth data, improving transparency and reliability.

Contribution

The paper presents a novel graph kernel-based method for detecting and explaining hallucinations in LLMs, enhancing interpretability and robustness in high-stakes applications.

Findings

Achieves competitive accuracy in hallucination detection

Provides contrastive, interpretable explanations

Works across open- and closed-domain tasks

Abstract

Large Language Models (LLMs) frequently generate hallucinations: statements that are syntactically plausible but lack factual grounding. This research presents KEA (Kernel-Enriched AI) Explain: a neurosymbolic framework that detects and explains such hallucinations by comparing knowledge graphs constructed from LLM outputs with ground truth data from Wikidata or contextual documents. Using graph kernels and semantic clustering, the method provides explanations for detected hallucinations, ensuring both robustness and interpretability. Our framework achieves competitive accuracy in detecting hallucinations across both open- and closed-domain tasks, and is able to generate contrastive explanations, enhancing transparency. This research advances the reliability of LLMs in high-stakes domains and provides a foundation for future work on precision improvements and multi-source knowledge integration.