weaviate/hard-questions-enronqa

这是一个包含138个问题的数据集，这些问题在使用cross encoder模型时能够达到recall @ 5的指标，但未能达到recall @ 1。数据集中的候选文档已经过预处理，使用了摘要推理来总结候选文档与查询的相关性，以减少输入长度并提高recall @ 1的指标。该数据集是从EnronQA数据集中抽样而来。

This dataset contains 138 questions where a cross encoder was able to achieve recall @ 5, but not recall @ 1. The candidate documents have been preprocessed with a summarization inference to summarize the relevance of the candidate document with respect to the query, which reduces the input length of the emails and significantly improves recall @ 1. The dataset is sampled from the EnronQA dataset.