2024/12/19 by Felix Friedrich, Simone Tedeschi, Friedrich, Felix +13 · 1 citation
Computer Science · #Computation and Language (cs.CL) #FOS: Computer and information sciences #Library Science and Information Systems #Natural Language Processing Techniques
paper · pdf · doi:10.48550/arxiv.2412.15035
openalex publication_date 2024/12/19 · openalex created_date 2024/12/21 · openalex updated_date 2026/07/28
Building safe Large Language Models (LLMs) across multiple languages is essential in ensuring both safe access and linguistic diversity. To this end, we conduct a large-scale, comprehensive safety evaluation of the current LLM landscape. For this purpose, we introduce M-ALERT, a multilingual benchmark that evaluates the safety of LLMs in five languages: English, French, German, Italian, and Spanish. M-ALERT includes 15k high-quality prompts per language, totaling 75k, with category-wise annotations. Our extensive experiments on 39 state-of-the-art LLMs highlight the importance of language-specific safety analysis, revealing that models often exhibit significant inconsistencies in safety across languages and categories. For instance, Llama3.2 shows high unsafety in category crimetax for Italian but remains safe in other languages. Similar inconsistencies can be observed across all models. In contrast, certain categories, such as substancecannabis and crimepropaganda, consistently trigger unsafe responses across models and languages. These findings underscore the need for robust multilingual safety practices in LLMs to ensure responsible usage across diverse communities.