QB2NQ is a dataset of natural-language questions automatically generated from Quiz Bowl (QB) questions using a series of linguistically motivated heuristic transformations. The resulting questions resemble real-world information-seeking queries while preserving the answerability of the original trivia questions, providing a scalable approach for training and evaluating open-domain question answering systems.
You Make… See the full description on the dataset page:
https://huggingface.co/datasets/TasnimKabir12/QB2NQ.
