SoftEDA: Rethinking Rule-Based Data Augmentation with Soft Labels

• DOI: 10.48550/arxiv.2402.05591
• Publication Date: 2024-02-08
• Authors: Juhwan Choi; Kyohoon Jin; Jun Ho Lee; Sangmin Song; Youngbin Kim

Rule-based text data augmentation is widely used for NLP tasks due to its simplicity. However, this method can potentially damage the original meaning of text, ultimately hurting performance model. To overcome limitation, we propose a straightforward technique applying soft labels augmented data. We conducted experiments across seven different classification and empirically demonstrated effectiveness our proposed approach. have publicly opened source code reproducibility.