Logo Lanfrica
  • Accueil
  • Atlas
  • Analyses
  • Documentation
  • Sign in

© 2026 Lanfrica. Tous droits réservés. Tous les droits d'auteur des ressources affichées sur le site Web Lanfrica appartiennent aux détenteurs de droits d'auteur d'origine, sauf indication contraire explicite.

Multilingual Biosecurity Safety Evaluation of Open-Weight Language Models: Evidence from African Languages

Domaine:

natural language processing

Type de record:

paper
Créateur:
Fat
Éditeur:
Zenodo
Hôte:avatar

This project evaluates the biosecurity related capabilities and safety behaviors of large language models across English, Hausa, and Swahili. By comparing model responses to the same biosecurity prompts in different languages, the study investigates whether safety safeguards, refusals, and scientific knowledge remain consistent across the different linguistic contexts. The goal is to identify potential multilingual safety gaps and contribute to more globally representative AI biosecurity evaluations.

 
 
 

Visit

doi.org

Languages

HausaSwahili

Licenses

info:eu-repo/semantics/openAccessCreative Commons Attribution 4.0 Internationalhttps://creativecommons.org/licenses/by/4.0/legalcode

Similaires

EthioLLM: Multilingual Large Language Models for Ethiopian Languages with Task EvaluationControlled Evaluation of Syntactic Knowledge in Multilingual Language ModelsControlled Evaluation of Syntactic Knowledge in Multilingual Language ModelsHealMed: Multilingual Evaluation of Large Language Models in MedicineResponsibly Building Multilingual Language Models for Hundreds of LanguagesContinually Adding New Languages to Multilingual Language Models

EthioLLM: Multilingual Large Language Models for Ethiopian Languages with Task Evaluation

Large language models (LLMs) have gained popularity recently due to their outstanding performance in

Controlled Evaluation of Syntactic Knowledge in Multilingual Language Models

Language models (LMs) are capable of acquiring elements of human-like syntactic knowledge. Targeted

Controlled Evaluation of Syntactic Knowledge in Multilingual Language Models

This benchmark is made up of "targeted syntactic evaluation tests for three low-resource languages (

HealMed: Multilingual Evaluation of Large Language Models in Medicine

We present HealMed, an expert-reviewed benchmark for multilingual evaluation of large language model

Responsibly Building Multilingual Language Models for Hundreds of Languages

Large Language Models (LLMs) have emerged as a transformative innovation in artificial intelligence,

Continually Adding New Languages to Multilingual Language Models

Multilingual language models are trained on a fixed set of languages, and to support new languages,