darkknight25/Adversarial_Machine_Learning_TextFooler_Dataset

Adversarial Machine Learning TextFooler Dataset Overview This dataset, adversarial_machine_learning_textfooler_dataset.jsonl, is designed for research and evaluation in adversarial machine learning, specifically focusing on text-based adversarial attacks. It contains pairs of original and adversarial text samples, primarily generated using the TextFooler attack method, along with other adversarial techniques such as Homoglyph, Number Substitution, Typo, Emoji, Semantic Shift, Paraphrase, and… See the full description on the dataset page: https://huggingface.co/datasets/darkknight25/Adversarial_Machine_Learning_TextFooler_Dataset.