Russian Texts Detoxification with Levenshtein Editing
arXiv:2204.13638v21 citationsh-index: 7
AI Analysis
This work addresses text detoxification for Russian language users, representing an incremental improvement over existing methods.
The paper tackled the problem of detoxifying Russian texts by using a two-step tagging-based model, achieving the best style transfer accuracy in the RUSSE Detox shared task.
Text detoxification is a style transfer task of creating neutral versions of toxic texts. In this paper, we use the concept of text editing to build a two-step tagging-based detoxification model using a parallel corpus of Russian texts. With this model, we achieved the best style transfer accuracy among all models in the RUSSE Detox shared task, surpassing larger sequence-to-sequence models.