This is the spoiler matching dataset as presented in Spoiler Detection as Semantic Text Matching. It consists of comments discussing episodes from various TV shows. Unlike other spoiler datasets, this dataset assigns an episode number (and show name) to each comment, enabling matching to specific episodes and very fine-grain spoiler detection. This dataset also includes an episode summary for each (show, episode) pair. For a given show, the task is to rank… See the full description on the dataset page:
https://huggingface.co/datasets/bobotran/spoiler-matching.
