Methods for Detoxification of Texts for the Russian Language

Abstract

We introduce the first study of automatic detoxification of Russian texts to combat offensive language. Such a kind of textual style transfer can be used, for instance, for processing toxic content in social media. While much work has been done for the English language in this field, it has never been solved for the Russian language yet. We test two types of models -unsupervised approach based on BERT architecture that performs local corrections and supervised approach based on pretrained language GPT-2 model -and compare them with several baselines. In addition, we describe evaluation setup providing training datasets and metrics for automatic evaluation. The results show that the tested approaches can be successfully used for detoxification, although there is room for improvement.

Knowledge Transfer Between Tasks and Languages in the Multi-task Encoder-agnostic Transformer-based Models

“…This dataset has 127,656 training samples, 93,342 validation samples, and 31,915 testing samples. For Russian toxic classification, we used the RuToxic dataset (Dementieva et al, 2021). This two-class dataset was collected from Dvach, a Russian anonymous imageboard.…”

RUSSE-2022: Findings of the First Russian Detoxification Shared Task Based on Parallel Corpora

“…Detoxification task is usually considered a variety of TST task from toxic to neutral style. There already exist unsupervised approaches to detoxification (Dementieva et al, 2021a;Dale et al, 2021) for the Russian and English languages. However, the output of these models is often of bad quality.…”