Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Yongheng, Liu, Xu, Zhou, Ruoxi, Chen, Qiguang, Fei, Hao, Lu, Wenpeng, Qin, Libo
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2505.19108
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866916757328887808
author Zhang, Yongheng
Liu, Xu
Zhou, Ruoxi
Chen, Qiguang
Fei, Hao
Lu, Wenpeng
Qin, Libo
author_facet Zhang, Yongheng
Liu, Xu
Zhou, Ruoxi
Chen, Qiguang
Fei, Hao
Lu, Wenpeng
Qin, Libo
contents Investigating hallucination issues in large language models (LLMs) within cross-lingual and cross-modal scenarios can greatly advance the large-scale deployment in real-world applications. Nevertheless, the current studies are limited to a single scenario, either cross-lingual or cross-modal, leaving a gap in the exploration of hallucinations in the joint cross-lingual and cross-modal scenarios. Motivated by this, we introduce a novel joint Cross-lingual and Cross-modal Hallucinations benchmark (CCHall) to fill this gap. Specifically, CCHall simultaneously incorporates both cross-lingual and cross-modal hallucination scenarios, which can be used to assess the cross-lingual and cross-modal capabilities of LLMs. Furthermore, we conduct a comprehensive evaluation on CCHall, exploring both mainstream open-source and closed-source LLMs. The experimental results highlight that current LLMs still struggle with CCHall. We hope CCHall can serve as a valuable resource to assess LLMs in joint cross-lingual and cross-modal scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2505_19108
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CCHall: A Novel Benchmark for Joint Cross-Lingual and Cross-Modal Hallucinations Detection in Large Language Models
Zhang, Yongheng
Liu, Xu
Zhou, Ruoxi
Chen, Qiguang
Fei, Hao
Lu, Wenpeng
Qin, Libo
Computation and Language
Artificial Intelligence
Investigating hallucination issues in large language models (LLMs) within cross-lingual and cross-modal scenarios can greatly advance the large-scale deployment in real-world applications. Nevertheless, the current studies are limited to a single scenario, either cross-lingual or cross-modal, leaving a gap in the exploration of hallucinations in the joint cross-lingual and cross-modal scenarios. Motivated by this, we introduce a novel joint Cross-lingual and Cross-modal Hallucinations benchmark (CCHall) to fill this gap. Specifically, CCHall simultaneously incorporates both cross-lingual and cross-modal hallucination scenarios, which can be used to assess the cross-lingual and cross-modal capabilities of LLMs. Furthermore, we conduct a comprehensive evaluation on CCHall, exploring both mainstream open-source and closed-source LLMs. The experimental results highlight that current LLMs still struggle with CCHall. We hope CCHall can serve as a valuable resource to assess LLMs in joint cross-lingual and cross-modal scenarios.
title CCHall: A Novel Benchmark for Joint Cross-Lingual and Cross-Modal Hallucinations Detection in Large Language Models
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2505.19108