Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Languages:
Spanish
Size:
10K - 100K
ArXiv:
Tags:
hate_speech