English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Text augmentation is one of the most effective techniques to solve thecritical problem of insufficient data in text classification. Existing textaugmentation methods achieve hopeful performance in few-shot text dataaugmentation. However, these methods usually lead to performance degenerationon public datasets due to poor quality augmentation instances. Our study showsthat even employing pre-trained language models, existing text augmentationmethods generate numerous low-quality instances and lead to the feature spaceshift problem in augmentation instances. However, we note that the pre-trainedlanguage model is good at finding low-quality instances provided that it hasbeen fine-tuned on the target dataset. To alleviate the feature space shift andperformance degeneration in existing text augmentation methods, we proposeBOOSTAUG, which reconsiders the role of the language model in text augmentationand emphasizes the augmentation instance filtering rather than generation. Weevaluate BOOSTAUG on both sentence-level text classification and aspect-basedsentiment classification. The experimental results on seven commonly used textclassification datasets show that our augmentation method obtainsstate-of-the-art performance. Moreover, BOOSTAUG is a flexible framework; werelease the code which can help improve existing augmentation methods.

Augmentor or Filter? Reconsider the Role of Pre-trained Language Model in Text Classification Augmentation