Manual Verbalizer Enrichment for Few-Shot Text Classification

TL;DR

This paper introduces MAVE, a method for enriching verbalizers in prompt-based few-shot text classification by leveraging word embeddings, achieving state-of-the-art results with minimal supervision.

Contribution

Proposes MAVE, a novel verbalizer enrichment technique using embedding neighborhood relations, and establishes a benchmarking procedure for few-shot document classification.

Findings

MAVE outperforms existing verbalizers in few-shot settings.

The approach is especially effective with very limited data.

Achieves state-of-the-art results with fewer resources.

Abstract

With the continuous development of pre-trained language models, prompt-based training becomes a well-adopted paradigm that drastically improves the exploitation of models for many natural language processing tasks. Prompting also shows great performance compared to traditional fine-tuning when adapted to zero-shot or few-shot scenarios where the number of annotated data is limited. In this framework, the role of verbalizers is essential, as an interpretation from masked word distributions into output predictions. In this work, we propose \acrshort{mave}, an approach for verbalizer construction by enrichment of class labels using neighborhood relation in the embedding space of words for the text classification task. In addition, we elaborate a benchmarking procedure to evaluate typical baselines of verbalizers for document classification in few-shot learning contexts. Our model achieves state-of-the-art results while using significantly fewer resources. We show that our approach is particularly effective in cases with extremely limited supervision data.