Large Language Models for Multilingual Previously Fact-Checked Claim Detection

2025-03-04

AI Summary

To address redundant fact-checking efforts caused by cross-lingual misinformation dissemination, this paper presents the first systematic study of large language models (LLMs) on multilingual debunked claim detection. We propose a cross-lingual semantic matching framework integrating prompt engineering, zero-/few-shot inference, and machine translation—specifically English-to-Chinese and low-resource-language-to-English translation. Experiments are conducted across seven state-of-the-art LLMs. We introduce the first benchmark dataset covering 20 languages. Results show that translating low-resource language inputs into English significantly improves detection performance (average +18.7%), revealing novel insights into LLMs’ cross-lingual semantic alignment capabilities; high-resource languages achieve high accuracy. This work provides a scalable, language-agnostic technical foundation for global multilingual fact-checking.

Technology Category

• Natural Language Processing: Fact-Checking / Misinformation Detection (NLP Focus)
• Machine Learning: Large Multimodal Models (LMMs)
• Computer Vision: Language and Vision

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

In our era of widespread false information, human fact-checkers often face the challenge of duplicating efforts when verifying claims that may have already been addressed in other countries or languages. As false information transcends linguistic boundaries, the ability to automatically detect previously fact-checked claims across languages has become an increasingly important task. This paper presents the first comprehensive evaluation of large language models (LLMs) for multilingual previously fact-checked claim detection. We assess seven LLMs across 20 languages in both monolingual and cross-lingual settings. Our results show that while LLMs perform well for high-resource languages, they struggle with low-resource languages. Moreover, translating original texts into English proved to be beneficial for low-resource languages. These findings highlight the potential of LLMs for multilingual previously fact-checked claim detection and provide a foundation for further research on this promising application of LLMs.

Problem

• Detecting previously fact-checked claims across multiple languages
• Evaluating large language models for multilingual claim detection
• Addressing challenges in low-resource language performance

Innovation

• Evaluates LLMs for multilingual claim detection
• Tests seven LLMs across 20 languages
• Uses translation to improve low-resource language performance