Perhaps I could obtain a dataset for NarrativeQA from you?

[DragonHou77]

I searched for the NarrativeQA dataset on the internet, but its composition is rather complex. I think it would be quite complicated to build a dataset for RAG. Additionally, I couldn't find the corpus and question-answer pairs of this dataset in your repository. Could you please provide a set of relevant files for the NarrativeQA dataset that you used?

[DragonHou77]

I downloaded the related projects of this dataset to the server, then executed the "download_stories.sh" script. Afterwards, I downloaded the crawled stories into a tmp document. However, these story files contain a large number of web HTML tags, and each file's corresponding qa pairs are in another csv file. So, what should I do next to build it into a corpus for RAG use and a set of question-answer pairs for evaluation?