orionweller/researchy-questions-b-only-low_level

该数据集包含三个主要配置：corpus、default和queries。corpus配置包含_id和text两个特征，test分割有63605个示例；default配置包含query-id、corpus-id和score三个特征，test分割有928921个示例；queries配置包含_id和text两个特征，test分割有282683个示例。此外，数据集的一个变体包括原始查询与GPT-4分解的各个子问题的组合，每个查询会针对每个子问题重复多次，并修改ID。

This dataset includes three configurations: corpus, default, and queries. The corpus configuration contains text data, the default configuration contains query IDs, corpus IDs, and scores, and the queries configuration also contains text data. The dataset variant includes the original query combined with individual subquestions from the GPT-4 decomposition, with each query duplicated multiple times to accommodate each subquestion and modified IDs.