nlp-udesa/hotpot_qa_3k

该数据集是一个用于验证的问答数据集，包含3000个示例，每个示例具有唯一标识符（id）、问题（question）、答案（answer）、类型（type）、级别（level）、支持事实（supporting_facts，包括标题和句子ID）和上下文（context，包括标题和句子序列）。数据集支持自然语言处理任务，如问答系统评估，但未在README中提供具体应用描述。This is a question answering dataset for validation purposes, containing 3000 instances. Each instance has a unique identifier (id), question, answer, type, level, supporting_facts (including title and sentence ID), and context (including title and sentence sequence). The dataset supports natural language processing tasks such as question answering system evaluation, but no specific application description is provided in the README.