HalluVerse25: Fine-grained Multilingual Benchmark Dataset for LLM Hallucinations

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Abdaljalil, Samir, Kurban, Hasan, Serpedin, Erchin
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912268749373440
author Abdaljalil, Samir
Kurban, Hasan
Serpedin, Erchin
author_facet Abdaljalil, Samir
Kurban, Hasan
Serpedin, Erchin
contents Large Language Models (LLMs) are increasingly used in various contexts, yet remain prone to generating non-factual content, commonly referred to as "hallucinations". The literature categorizes hallucinations into several types, including entity-level, relation-level, and sentence-level hallucinations. However, existing hallucination datasets often fail to capture fine-grained hallucinations in multilingual settings. In this work, we introduce HalluVerse25, a multilingual LLM hallucination dataset that categorizes fine-grained hallucinations in English, Arabic, and Turkish. Our dataset construction pipeline uses an LLM to inject hallucinations into factual biographical sentences, followed by a rigorous human annotation process to ensure data quality. We evaluate several LLMs on HalluVerse25, providing valuable insights into how proprietary models perform in detecting LLM-generated hallucinations across different contexts.
format Preprint
id arxiv_https___arxiv_org_abs_2503_07833
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle HalluVerse25: Fine-grained Multilingual Benchmark Dataset for LLM Hallucinations
Abdaljalil, Samir
Kurban, Hasan
Serpedin, Erchin
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) are increasingly used in various contexts, yet remain prone to generating non-factual content, commonly referred to as "hallucinations". The literature categorizes hallucinations into several types, including entity-level, relation-level, and sentence-level hallucinations. However, existing hallucination datasets often fail to capture fine-grained hallucinations in multilingual settings. In this work, we introduce HalluVerse25, a multilingual LLM hallucination dataset that categorizes fine-grained hallucinations in English, Arabic, and Turkish. Our dataset construction pipeline uses an LLM to inject hallucinations into factual biographical sentences, followed by a rigorous human annotation process to ensure data quality. We evaluate several LLMs on HalluVerse25, providing valuable insights into how proprietary models perform in detecting LLM-generated hallucinations across different contexts.
title HalluVerse25: Fine-grained Multilingual Benchmark Dataset for LLM Hallucinations
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2503.07833