Prompt-Based Metric Learning for Few-Shot NER

TL;DR

This paper introduces a prompt-based metric learning approach for few-shot NER that enhances label semantics and combines multiple representations, achieving state-of-the-art results across various settings.

Contribution

It proposes a novel prompt schema and architecture to improve metric learning for few-shot NER, significantly outperforming previous methods.

Findings

Achieves state-of-the-art results in 16 out of 18 settings.

Substantially outperforms previous SOTA with up to 34.51% relative F1 gain.

Demonstrates effectiveness of prompt schemas in few-shot NER.

Abstract

Few-shot named entity recognition (NER) targets generalizing to unseen labels and/or domains with few labeled examples. Existing metric learning methods compute token-level similarities between query and support sets, but are not able to fully incorporate label semantics into modeling. To address this issue, we propose a simple method to largely improve metric learning for NER: 1) multiple prompt schemas are designed to enhance label semantics; 2) we propose a novel architecture to effectively combine multiple prompt-based representations. Empirically, our method achieves new state-of-the-art (SOTA) results under 16 of the 18 considered settings, substantially outperforming the previous SOTA by an average of 8.84% and a maximum of 34.51% in relative gains of micro F1. Our code is available at https://github.com/AChen-qaq/ProML.