sarvamai/trivia-qa-indic

Indic TriviaQA数据集是TriviaQA阅读理解数据集的多语言版本，从英语翻译成10种印度语言。该版本遵循原始数据集的无上下文格式，包含由琐事爱好者编写的问题-答案对的验证集和测试集的翻译。数据集涵盖的语言包括孟加拉语、古吉拉特语、印地语、卡纳达语、马拉地语、马拉雅拉姆语、奥里亚语、旁遮普语、泰米尔语和泰卢固语。每个示例包含问题文本、答案（包括答案别名和标准化的别名）和语言代码。验证集和测试集分别包含约18k和17k个问题。

The Indic TriviaQA dataset is a multilingual version of the TriviaQA Reading Comprehension dataset, translated from English into 10 Indian languages. It contains translations of the validation and test sets of question-answer pairs authored by trivia enthusiasts and independently gathered evidence documents. The dataset includes translations in Bengali, Gujarati, Hindi, Kannada, Marathi, Malayalam, Oriya, Punjabi, Tamil, and Telugu. Each example contains a question text, answer (including a list of acceptable answers and normalized aliases), and ISO 639-1 language code. The validation and test sets contain approximately 18k and 17k questions per language, respectively.