README.md

metadata

task_categories:
  - text-classification
language:
  - en
pretty_name: Dynamic-Escape

perturbed toxic text dataset (from Jigsaw dataset), managed by the types of perturbations. There are 9 types of perturbations:

1. insert
2. repeat
3. maskword
4. homoglyph
5. swap
6. remove
7. abbrs./slangs
8. distract words
9. distract sentences

You also can access to Github: https://github.com/khk-abc/DynEscape---Dynamic-Escape

The paper has been published. Please cite the following paper if you would like to use this dataset.

@article{kang2026developing,
  title={Developing continuous toxicity detection against increasing types of perturbed toxic text},
  author={Kang, Hankun and Chen, Jianhao and Li, Yongqi and Miao, Xin and Xu, Mayi and Zhou, Shen and Zhong, Ming and Zhu, Yuanyuan and Qian, Tieyun},
  journal={Information Processing \& Management},
  volume={63},
  number={2},
  pages={104470},
  year={2026},
  publisher={Elsevier}
}