LLaMAX/BenchMAX_Science
收藏Hugging Face2025-02-12 更新2025-02-15 收录
下载链接：
https://hf-mirror.com/datasets/LLaMAX/BenchMAX_Science
下载链接
链接失效反馈官方服务：
资源简介：
BenchMAX_Science是一个多语言数据集，来源于GPQA数据集。该数据集将原始的英语数据集扩展到了16种非英语语言，包括阿拉伯语、孟加拉语、中文、捷克语、英语、法语、德语、匈牙利语、日语、韩语、塞尔维亚语、西班牙语、斯瓦希里语、泰卢固语、泰语、俄语和越南语。数据集用于文本生成任务，经过谷歌翻译和母语人士的后期编辑，规模在1K到10K之间。
BenchMAX_Science is a multilingual dataset based on the GPQA dataset. The dataset extends the original English dataset to 16 non-English languages, including Arabic, Bengali, Chinese, Czech, English, French, German, Hungarian, Japanese, Korean, Serbian, Spanish, Swahili, Telugu, Thai, Russian, and Vietnamese. It is used for text generation tasks, has been translated by Google Translate and post-edited by native speakers, and is sized between 1K and 10K.
提供机构：
LLaMAX
