GASP is a dataset composed by a list of cited abstracts associated with the corresponding source abstract. The dataset is composed by a training set of 100000 elements, a test set and a validation set of 10000 each. The goal is to generate a paper abstract given cited paper's abstracts and model the human creativity behind the process.

Source: https://github.com/ART-Group-it/GASP

Papers


Paper Code Results Date Stars

Dataset Loaders


Tasks


Similar Datasets


License


  • Unknown

Modalities


Languages