Russian Toxicity Classifier

🚀 Russian Toxic Comments Classifier

A BERT-based classifier for detecting toxic comments in Russian, fine-tuned from a conversational model.

🚀 Quick Start

This BERT-based classifier is fine-tuned from Conversational Rubert. It is trained on a merged dataset of Russian toxic comments from 2ch.hk and ok.ru.

The combined datasets were shuffled and split into training, development, and test sets in an 80-10-10 ratio. The following metrics were obtained from the test dataset:

💻 Usage Examples

Basic Usage

from transformers import BertTokenizer, BertForSequenceClassification

# load tokenizer and model weights
tokenizer = BertTokenizer.from_pretrained('s-nlp/russian_toxicity_classifier')
model = BertForSequenceClassification.from_pretrained('s-nlp/russian_toxicity_classifier')

# prepare the input
batch = tokenizer.encode('ты супер', return_tensors='pt')

# inference
model(batch)

📚 Documentation

Citation

To acknowledge our work, please use the following citation:

@article{dementieva2022russe,
  title={RUSSE-2022: Findings of the First Russian Detoxification Shared Task Based on Parallel Corpora},
  author={Dementieva, Daryna and Logacheva, Varvara and Nikishina, Irina and Fenogenova, Alena and Dale, David and Krotova, Irina and Semenov, Nikita and Shavrina, Tatiana and Panchenko, Alexander}
}

📄 License

This model is licensed under the OpenRAIL++ License, which supports the development of various technologies—both industrial and academic—that serve the public good.