Edit model card

This is the detoxification baseline model trained on the train part of "RUSSE 2022: Russian Text Detoxification Based on Parallel Corpora" competition. The source sentences are Russian toxic messages from Odnoklassniki, Pikabu, and Twitter platforms. The base model is ruT5.

How to use

from transformers import T5ForConditionalGeneration, AutoTokenizer

base_model_name = 'ai-forever/ruT5-base'
model_name = 's-nlp/ruT5-base-detox'

tokenizer = AutoTokenizer.from_pretrained(base_model_name)
model = T5ForConditionalGeneration.from_pretrained(model_name)

input_ids = tokenizer.encode('Это полная хуйня!', return_tensors='pt')
output_ids = model.generate(input_ids, max_length=50, num_return_sequences=1)
output_text = tokenizer.decode(output_ids[0], skip_special_tokens=True)
print(output_text)
# Это полный бред!

Citation

@article{dementievarusse,
  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.

Downloads last month
41
Safetensors
Model size
223M params
Tensor type
F32
·
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for s-nlp/ruT5-base-detox

Finetuned
(12)
this model

Dataset used to train s-nlp/ruT5-base-detox