license: openrail
You can download this Dataset just like this:
data_files = {"train": "data_nli_train_df.csv",
"validation": "data_nli_val_df.csv",
"test": "data_nli_test_df.csv"}
dataset = load_dataset("muhammadravi251001/idk_mrc_nli_ner", data_files=data_files)
This is some modification from IDK-MRC dataset to IDK-MRC-NLI dataset. By convert QAS dataset to NLI dataset. You can find the original IDK-MRC in this link: https://huggingface.co/datasets/rifkiaputri/idk-mrc
@inproceedings{putri-oh-2022-idk, title = "{IDK}-{MRC}: Unanswerable Questions for {I}ndonesian Machine Reading Comprehension", author = "Putri, Rifki Afina and Oh, Alice", booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing", month = dec, year = "2022", address = "Abu Dhabi, United Arab Emirates", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.emnlp-main.465", pages = "6918--6933", }