bzantium/MMMLU

MMMLU是一个广泛认可的用于评估AI模型所获得通用知识的基准，包含57个不同类别的话题，从基础知识的水平到像法律、物理、历史和计算机科学这样的高级专业主题。该数据集将MMLU的测试集翻译成了14种语言，使用专业的人工翻译以保证翻译的准确性，特别适用于像约鲁巴语这样的低资源语言。这项工作体现了提高AI模型多语言能力的承诺，确保它们在不同语言中准确执行，尤其是对代表性不足的社区。通过优先考虑高质量的翻译，旨在使AI技术对全球用户更加包容和有效。

The MMMLU is a widely recognized benchmark for evaluating the general knowledge attained by AI models, covering topics from 57 different categories ranging from elementary-level knowledge to advanced professional subjects such as law, physics, history, and computer science. The test set of MMLU has been translated into 14 languages using professional human translators to ensure translation accuracy, especially for low-resource languages like Yoruba. This effort reflects a commitment to improving the multilingual capabilities of AI models, ensuring accurate performance across languages and particularly for underrepresented communities. By prioritizing high-quality translations, the aim is to make AI technology more inclusive and effective for users worldwide.