# Improving Results on Russian Sentiment Datasets

arXiv: https://arxiv.org/abs/2007.14310 · HTML: https://mlsanity.com/paper/2007.14310/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 28 Jul 2020.

## Built on by (most cited first, 5 of 5)

- [RuSentNE-2023: Evaluating Entity-Oriented Sentiment Analysis on Russian News Texts](https://mlsanity.com/paper/2305.17679/) (2023)
- [Large Language Models in Targeted Sentiment Analysis](https://mlsanity.com/paper/2404.12342/) (2024)
- [Transfer Learning for Improving Results on Russian Sentiment Datasets](https://mlsanity.com/paper/2107.02499/) (2021)
- [Current Landscape of the Russian Sentiment Corpora](https://mlsanity.com/paper/2106.14434/) (2021)
- [RuOpinionNE-2024: Extraction of Opinion Tuples from Russian News Texts](https://mlsanity.com/paper/2504.06947/) (2025)

## Abstract (arXiv)

In this study, we test standard neural network architectures (CNN, LSTM, BiLSTM) and recently appeared BERT architectures on previous Russian sentiment evaluation datasets. We compare two variants of Russian BERT and show that for all sentiment tasks in this study the conversational variant of Russian BERT performs better. The best results were achieved by BERT-NLI model, which treats sentiment classification tasks as a natural language inference task. On one of the datasets, this model practically achieves the human level.