SoftEDA：重新思考基于规则的软标签数据增强
SoftEDA: Rethinking Rule-Based Data Augmentation with Soft Labels
AI总结：
针对基于规则的数据增强可能损害文本原义的问题，提出软标签技术，在七个分类任务上验证其有效性，并公开源码。
摘要：
基于规则的文本数据增强因其简单性而被广泛应用于自然语言处理任务。然而，这种方法可能损害文本的原始含义，最终影响模型的性能。为了克服这一局限，我们提出了一种将软标签应用于增强数据的直接技术。我们在七个不同的分类任务上进行了实验，并实证证明了所提出方法的有效性。我们已公开源代码以确保可复现性。
英文摘要：
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 the text, ultimately hurting the performance of the model. To overcome this limitation, we propose a straightforward technique for applying soft labels to augmented data. We conducted experiments across seven different classification tasks and empirically demonstrated the effectiveness of our proposed approach. We have publicly opened our source code for reproducibility.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
