English

BootAug: Boosting Text Augmentation via Hybrid Instance Filtering Framework

Computation and Language 2024-04-02 v2

Abstract

Text augmentation is an effective technique for addressing the problem of insufficient data in natural language processing. However, existing text augmentation methods tend to focus on few-shot scenarios and usually perform poorly on large public datasets. Our research indicates that existing augmentation methods often generate instances with shifted feature spaces, which leads to a drop in performance on the augmented data (for example, EDA generally loses 2%\approx 2\% in aspect-based sentiment classification). To address this problem, we propose a hybrid instance-filtering framework (BootAug) based on pre-trained language models that can maintain a similar feature space with natural datasets. BootAug is transferable to existing text augmentation methods (such as synonym substitution and back translation) and significantly improves the augmentation performance by 23%\approx 2-3\% in classification accuracy. Our experimental results on three classification tasks and nine public datasets show that BootAug addresses the performance drop problem and outperforms state-of-the-art text augmentation methods. Additionally, we release the code to help improve existing augmentation methods on large datasets.

Keywords

Cite

@article{arxiv.2210.02941,
  title  = {BootAug: Boosting Text Augmentation via Hybrid Instance Filtering Framework},
  author = {Heng Yang and Ke Li},
  journal= {arXiv preprint arXiv:2210.02941},
  year   = {2024}
}

Comments

Source code and examples: https://github.com/yangheng95/BoostTextAugmentation

R2 v1 2026-06-28T02:56:00.873Z