# BioNLP2021 dataset (Task2)
___

Data fields:
* text (str): source text; Section and Article (train_mul subset only) are separated by <SAS> ; Single Documents  are separated by <DOC> ; Sentences  are separated by <SS>
* summ_abs, summ_ext (str): abstractive and extractive summarization, whose Sentences  are separated by <SS>
* question (str): question, whose Sentences  are separated by <SS>
* key (str): key in the origin dataset (for submitting)