The CSQA dataset is a widely used benchmark dataset for conversational KBQA, consisting of around 200K dialogues where training set, validation set and testing set contain 153K, 16K and 28K dialogues, respectively.
BibTex:
Before browse our site, please accept our cookies policy
