metadata
license: cc-by-4.0
language:
- en
configs:
- config_name: MNLI
data_files:
- split: train
path: MNLI/mnli_train_textflint.csv
- split: val_matched
path: MNLI/mnli_val_matched_textflint.csv
- split: val_mismatched
path: MNLI/mnli_val_mismatched_textflint.csv
- config_name: QNLI
data_files:
- split: train
path: QNLI/qnli_train_textflint.csv
- split: val
path: QNLI/qnli_val_textflint.csv
- config_name: QQP
data_files:
- split: train
path: QQP/qqp_train_textflint.csv
- split: val
path: QQP/qqp_val_textflint.csv
- config_name: SST2
data_files:
- split: train
path: SST2/sst2_dev_textflint.csv
- split: val
path: SST2/sst2_train_textflint.csv
GLUE+typo tasks used in "Tokenization is Sensitive to Language Variation paper", Arxiv link
@article{wegmann2025tokenization,
title={Tokenization is Sensitive to Language Variation},
author={Wegmann, Anna and Nguyen, Dong and Jurgens, David},
journal={arXiv preprint arXiv:2502.15343},
year={2025}
}