QaNER: Prompting Question Answering Models for Few-shot Named Entity Recognition

TL;DR

QaNER introduces a prompt-based question answering approach for few-shot NER that improves efficiency, robustness, and zero-shot performance compared to previous methods.

Contribution

It proposes a novel QA-based prompt method for NER, addressing limitations of prior prompt-based approaches in efficiency and zero-shot ability.

Findings

Faster inference compared to previous methods.

More robust to prompt quality and hyper-parameters.

Significantly better performance in low-resource and zero-shot scenarios.

Abstract

Recently, prompt-based learning for pre-trained language models has succeeded in few-shot Named Entity Recognition (NER) by exploiting prompts as task guidance to increase label efficiency. However, previous prompt-based methods for few-shot NER have limitations such as a higher computational complexity, poor zero-shot ability, requiring manual prompt engineering, or lack of prompt robustness. In this work, we address these shortcomings by proposing a new prompt-based learning NER method with Question Answering (QA), called QaNER. Our approach includes 1) a refined strategy for converting NER problems into the QA formulation; 2) NER prompt generation for QA models; 3) prompt-based tuning with QA models on a few annotated NER examples; 4) zero-shot NER by prompting the QA model. Comparing the proposed approach with previous methods, QaNER is faster at inference, insensitive to the prompt quality, and robust to hyper-parameters, as well as demonstrating significantly better low-resource performance and zero-shot capability.