Label-template based Few-Shot Text Classification with Contrastive Learning

TL;DR

This paper introduces a label-template based few-shot text classification framework that leverages label semantics and contrastive learning to improve discriminative text representations, outperforming existing methods.

Contribution

It proposes a novel framework integrating label templates and supervised contrastive learning with an attention mechanism for enhanced few-shot text classification.

Findings

Achieves significant performance improvements over state-of-the-art models.

Effectively utilizes label semantics to guide representation learning.

Demonstrates robustness across four different datasets.

Abstract

As an algorithmic framework for learning to learn, meta-learning provides a promising solution for few-shot text classification. However, most existing research fail to give enough attention to class labels. Traditional basic framework building meta-learner based on prototype networks heavily relies on inter-class variance, and it is easily influenced by noise. To address these limitations, we proposes a simple and effective few-shot text classification framework. In particular, the corresponding label templates are embed into input sentences to fully utilize the potential value of class labels, guiding the pre-trained model to generate more discriminative text representations through the semantic information conveyed by labels. With the continuous influence of label semantics, supervised contrastive learning is utilized to model the interaction information between support samples and query samples. Furthermore, the averaging mechanism is replaced with an attention mechanism to highlight vital semantic information. To verify the proposed scheme, four typical datasets are employed to assess the performance of different methods. Experimental results demonstrate that our method achieves substantial performance enhancements and outperforms existing state-of-the-art models on few-shot text classification tasks.