论文标题
改善俄罗斯情感数据集的结果
Improving Results on Russian Sentiment Datasets
论文作者
论文摘要
在这项研究中，我们测试了标准的神经网络体系结构（CNN，LSTM，Bilstm），并最近出现在先前的俄罗斯情绪评估数据集中。我们比较了俄罗斯伯特的两个变体，并表明，对于本研究的所有情感任务，俄罗斯伯特的对话变体的表现更好。最好的结果是通过Bert-NLI模型实现的，该模型将情感分类任务视为自然语言推理任务。在其中一个数据集上，该模型实际上实现了人类水平。
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.
