Few-shot Text Classification with Dual Contrastive Consistency

TL;DR

This paper introduces FTCC, a novel approach combining supervised contrastive learning and consistency regularization to improve few-shot text classification with pre-trained language models, outperforming existing methods.

Contribution

The paper proposes a new contrastive consistency method and a training framework that enhances few-shot text classification performance and robustness.

Findings

Outperforms state-of-the-art methods on four datasets.

Demonstrates improved robustness in few-shot classification.

Effectively utilizes unlabeled data through consistency regularization.

Abstract

In this paper, we explore how to utilize pre-trained language model to perform few-shot text classification where only a few annotated examples are given for each class. Since using traditional cross-entropy loss to fine-tune language model under this scenario causes serious overfitting and leads to sub-optimal generalization of model, we adopt supervised contrastive learning on few labeled data and consistency-regularization on vast unlabeled data. Moreover, we propose a novel contrastive consistency to further boost model performance and refine sentence representation. After conducting extensive experiments on four datasets, we demonstrate that our model (FTCC) can outperform state-of-the-art methods and has better robustness.