KEA Explain: Explanations of Hallucinations using Graph Kernel Analysis

Reilly Haskins, Benjamin Adams

Keywords: large language models, hallucination, knowledge graph, neurosymbolic, graph kernel, explainability

TL;DR: A neurosymbolic framework that detects and explains LLM hallucinations using knowledge graphs, graph kernels, and semantic clustering for high accuracy and transparency.

Track: Neurosymbolic Methods for Trustworthy and Interpretable AI

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.

Paper Type: Long Paper

Software: https://github.com/Reih02/hallucination_explanation_graph_kernel_analysis

Submission Number: 24