Text Augmentation in a Multi-Task View
This is an incremental improvement for text classification tasks, addressing robustness in data augmentation.
The paper tackles the problem of limited effectiveness in traditional data augmentation by proposing a multi-task view (MTV) that treats original and augmented data as separate tasks, allowing stronger augmentation. It shows that MTV leads to higher and more robust performance improvements on three benchmark text classification datasets compared to traditional methods.
Traditional data augmentation aims to increase the coverage of the input distribution by generating augmented examples that strongly resemble original samples in an online fashion where augmented examples dominate training. In this paper, we propose an alternative perspective -- a multi-task view (MTV) of data augmentation -- in which the primary task trains on original examples and the auxiliary task trains on augmented examples. In MTV data augmentation, both original and augmented samples are weighted substantively during training, relaxing the constraint that augmented examples must resemble original data and thereby allowing us to apply stronger levels of augmentation. In empirical experiments using four common data augmentation techniques on three benchmark text classification datasets, we find that the MTV leads to higher and more robust performance improvements than traditional augmentation.