This is the dataset for COVID-19 document type screening. It is composed of: - Epistemonikos train dataset - CORD-19 test dataset adapted for Evidence Based Medicine domain - XLNET model fine-tuned on Epistemonikos dataset. - BioBERT model fine-tuned on Epistemonikos dataset. Epistemonikos XLNET models fine-tuned on Cord-19: - Episte-XLNET fine-tuned with random sampling strategy. - Episte-XLNET fine-tuned with data augmentation strategy. - Episte-XLNET fine-tuned with uncertainty sampling strategy (iteration 1 and 2). Scripts to run experiments can be found at: https://github.com/afcarvallo/covid_19_document_type_screening