MultiHal: Multilingual Dataset for Knowledge-Graph Grounded Evaluation of LLM Hallucinations

TL;DR

MultiHal introduces a multilingual, knowledge graph-based benchmark for evaluating and mitigating hallucinations in large language models, leveraging structured factual data across multiple languages and improving factuality assessment.

Contribution

It creates a novel multilingual, multihop knowledge graph benchmark for LLM hallucination evaluation, filling gaps in existing datasets by integrating structured factual resources.

Findings

Improved semantic similarity scores by 0.12 to 0.36 points

Enhanced NLI entailment accuracy by 0.16 to 0.36 points

Increased hallucination detection performance by 0.29 to 0.42 points

Abstract

Large Language Models (LLMs) have inherent limitations of faithfulness and factuality, commonly referred to as hallucinations. Several benchmarks have been developed that provide a test bed for factuality evaluation within the context of English-centric datasets, while relying on supplementary informative context like web links or text passages but ignoring the available structured factual resources. To this end, Knowledge Graphs (KGs) have been identified as a useful aid for hallucination mitigation, as they provide a structured way to represent the facts about entities and their relations with minimal linguistic overhead. We bridge the lack of KG paths and multilinguality for factual language modeling within the existing hallucination evaluation benchmarks and propose a KG-based multilingual, multihop benchmark called MultiHal framed for generative text evaluation. As part of our data collection pipeline, we mined 140k KG-paths from open-domain KGs, from which we pruned noisy KG-paths, curating a high-quality subset of 25.9k. Our baseline evaluation shows an absolute scale improvement by approximately 0.12 to 0.36 points for the semantic similarity score, 0.16 to 0.36 for NLI entailment and 0.29 to 0.42 for hallucination detection in KG-RAG over vanilla QA across multiple languages and multiple models, demonstrating the potential of KG integration. We anticipate MultiHal will foster future research towards several graph-based hallucination mitigation and fact-checking tasks.