Adversarial Demonstration Learning for Low-resource NER Using Dual Similarity

TL;DR

This paper introduces a dual similarity-based demonstration selection and adversarial training approach to improve low-resource NER performance, addressing issues in demonstration relevance and model referencing.

Contribution

It proposes a novel dual similarity method for selecting demonstrations and an adversarial training framework to enhance model referencing in low-resource NER.

Findings

Outperforms existing methods in low-resource NER tasks

Dual similarity selection improves demonstration relevance

Adversarial training enhances model referencing ability

Abstract

We study the problem of named entity recognition (NER) based on demonstration learning in low-resource scenarios. We identify two issues in demonstration construction and model training. Firstly, existing methods for selecting demonstration examples primarily rely on semantic similarity; We show that feature similarity can provide significant performance improvement. Secondly, we show that the NER tagger's ability to reference demonstration examples is generally inadequate. We propose a demonstration and training approach that effectively addresses these issues. For the first issue, we propose to select examples by dual similarity, which comprises both semantic similarity and feature similarity. For the second issue, we propose to train an NER model with adversarial demonstration such that the model is forced to refer to the demonstrations when performing the tagging task. We conduct comprehensive experiments in low-resource NER tasks, and the results demonstrate that our method outperforms a range of methods.