Knowledge-Aware Meta-learning for Low-Resource Text Classification

TL;DR

This paper introduces KGML, a knowledge-aware meta-learning approach that enhances low-resource text classification by incorporating external knowledge bases, improving generalization to new tasks.

Contribution

It proposes a novel method that integrates sentence-specific knowledge graphs into meta-learning for better low-resource text classification.

Findings

KGML outperforms baseline models on three datasets.

Effective in both supervised and unsupervised adaptation.

Enhances generalization to unseen tasks.

Abstract

Meta-learning has achieved great success in leveraging the historical learned knowledge to facilitate the learning process of the new task. However, merely learning the knowledge from the historical tasks, adopted by current meta-learning algorithms, may not generalize well to testing tasks when they are not well-supported by training tasks. This paper studies a low-resource text classification problem and bridges the gap between meta-training and meta-testing tasks by leveraging the external knowledge bases. Specifically, we propose KGML to introduce additional representation for each sentence learned from the extracted sentence-specific knowledge graph. The extensive experiments on three datasets demonstrate the effectiveness of KGML under both supervised adaptation and unsupervised adaptation settings.