FlipDA: Effective and Robust Data Augmentation for Few-Shot Learning
Most previous methods for text data augmentation are limited to simple tasks and weak baselines. We explore data augmentation on hard tasks (i.e., few-shot natural language understanding) and strong baselines (i.e., pretrained models with over one billion parameters). Under this setting, we reproduced a large number of previous augmentation methods and found that these methods bring marginal gains at best and sometimes degrade the performance much. To address this challenge, we propose a novel data augmentation method FlipDA that jointly uses a generative model and a classifier to generate label-flipped data. Central to the idea of FlipDA is the discovery that generating label-flipped data is more crucial to the performance than generating label-preserved data. Experiments show that FlipDA achieves a good tradeoff between effectiveness and robustness -- it substantially improves many tasks while not negatively affecting the others.
Code (1)
Tasks
Data AugmentationFew-Shot LearningNatural Language UnderstandingSimilar Papers 제목 키워드 기반
FlipDA: Effective and Robust Data Augmentation for Few-Shot Learning
Most previous methods for text data augmentation are limited to simple tasks and weak baselines. We explore data augmentation on hard tasks (i.e., few-shot natural language understanding) and strong baselines (i.e., pret…
Data AugmentationFew-Shot LearningNatural Language UnderstandingPromptDA: Label-guided Data Augmentation for Prompt-based Few-shot Learners
Recent advances in large pre-trained language models (PLMs) lead to impressive gains in natural language understanding (NLU) tasks with task-specific fine-tuning. However, directly fine-tuning PLMs heavily relies on suff…
Data AugmentationFew-Shot LearningFew-Shot Text ClassificationNatural Language Understanding+2Label-guided Data Augmentation for Prompt-based Few Shot Learners
Recent advances on large pre-trained language models (PLMs) lead impressive gains on many natural language understanding (NLU) tasks with task-specific fine-tuning. However, direct fine-tuning PLMs heavily rely on large …
Data AugmentationFew-Shot LearningFew-Shot Text ClassificationNatural Language Understanding+2Diversity Helps: Unsupervised Few-shot Learning via Distribution Shift-based Data Augmentation
Few-shot learning aims to learn a new concept when only a few training examples are available, which has been extensively explored in recent years. However, most of the current works heavily rely on a large-scale labeled…
Data AugmentationDiversityFew-Shot LearningUnsupervised Few-Shot Image Classification+1Few-Shot Knowledge Graph Completion with Data Fusion and Augmentation
This paper addresses the few-shot knowledge graph completion problem, which aims to infer facts for long-tail distributed relations for completing knowledge graphs. The few-shot knowledge graph completion task confronts …
Few-Shot LearningKnowledge Graph CompletionKnowledge GraphsWorld Knowledge