Show Less, Instruct More: Enriching Prompts with Definitions and Guidelines for Zero-Shot NER

TL;DR

This paper introduces SLIMER, a novel approach that enhances zero-shot NER by enriching prompts with definitions and guidelines, enabling better generalization to unseen entity types with fewer examples.

Contribution

SLIMER leverages enriched prompts with definitions and guidelines to improve zero-shot NER performance on unseen entity tags, offering a more robust and efficient learning method.

Findings

Enriched prompts improve zero-shot NER accuracy on unseen entities.

SLIMER performs comparably to state-of-the-art methods in out-of-domain settings.

Fewer examples suffice for effective learning with enriched prompts.

Abstract

Recently, several specialized instruction-tuned Large Language Models (LLMs) for Named Entity Recognition (NER) have emerged. Compared to traditional NER approaches, these models have demonstrated strong generalization capabilities. Existing LLMs primarily focus on addressing zero-shot NER on Out-of-Domain inputs, while fine-tuning on an extensive number of entity classes that often highly or completely overlap with test sets. In this work instead, we propose SLIMER, an approach designed to tackle never-seen-before entity tags by instructing the model on fewer examples, and by leveraging a prompt enriched with definition and guidelines. Experiments demonstrate that definition and guidelines yield better performance, faster and more robust learning, particularly when labelling unseen named entities. Furthermore, SLIMER performs comparably to state-of-the-art approaches in out-of-domain zero-shot NER, while being trained in a more fair, though certainly more challenging, setting.