When Low Resource NLP Meets Unsupervised Language Model: Meta-pretraining Then Meta-learning for Few-shot Text Classification
This addresses data deficiency and adaptation to unseen classes in NLP, but it is incremental as it builds on existing meta-learning and pretraining methods.
The paper tackles few-shot text classification by combining meta-pretraining with meta-learning, achieving state-of-the-art performance on a sentiment classification dataset.
Text classification tends to be difficult when data are deficient or when it is required to adapt to unseen classes. In such challenging scenarios, recent studies have often used meta-learning to simulate the few-shot task, thus negating implicit common linguistic features across tasks. This paper addresses such problems using meta-learning and unsupervised language models. Our approach is based on the insight that having a good generalization from a few examples relies on both a generic model initialization and an effective strategy for adapting this model to newly arising tasks. We show that our approach is not only simple but also produces a state-of-the-art performance on a well-studied sentiment classification dataset. It can thus be further suggested that pretraining could be a promising solution for few-shot learning of many other NLP tasks. The code and the dataset to replicate the experiments are made available at https://github.com/zxlzr/FewShotNLP.