Abstract:Most existing NER methods rely on extensive labeled data for model training, which struggles in the low-resource scenarios with limited training data. Recently, prompt-tuning methods for pre-trained language models have achieved remarkable performance in few-shot learning by exploiting prompts as task guidance to reduce the gap between training progress and downstream tuning. Inspired by prompt learning, we propose a novel lightweight generative framework with prompt-guided attention for low-resource NER (LightNER). Specifically, we construct the semantic-aware answer space of entity categories for prompt learning to generate the entity span sequence and entity categories without any label-specific classifiers. We further propose prompt-guided attention by incorporating continuous prompts into the self-attention layer to re-modulate the attention and adapt pre-trained weights. Note that we only tune those continuous prompts with the whole parameter of the pre-trained language model fixed, thus, making our approach lightweight and flexible for low-resource scenarios and can better transfer knowledge across domains. Experimental results show that LightNER can obtain comparable performance in the standard supervised setting and outperform strong baselines in low-resource settings by tuning only a small part of the parameters.

Prompt-Based Data Augmentation Framework for Few-Shot Named Entity Recognition

Enhancing Few-shot NER with Prompt Ordering based Data Augmentation

Low-Resource NER by Data Augmentation with Prompting

Prompt-Based Self-training Framework for Few-Shot Named Entity Recognition.

RoPDA: Robust Prompt-based Data Augmentation for Low-Resource Named Entity Recognition.

Focusing, Bridging and Prompting for Few-shot Nested Named Entity Recognition

LLM-DA: Data Augmentation via Large Language Models for Few-Shot Named Entity Recognition

Boosting LLMS with Ontology-Aware Prompt for Ner Data Augmentation

Structure and Label Constrained Data Augmentation for Cross-domain Few-shot NER

Data Augmentation for Cross-Domain Named Entity Recognition

PromptNER: A Prompting Method for Few-shot Named Entity Recognition via k Nearest Neighbor Search

Graph Propagation based Data Augmentation for Named Entity Recognition

LightNER: A Lightweight Generative Framework with Prompt-guided Attention for Low-resource NER

Guidance-Based Prompt Data Augmentation in Specialized Domains for Named Entity Recognition

Evaluating Named Entity Recognition Using Few-Shot Prompting with Large Language Models

Improving Low-resource Named Entity Recognition with Graph Propagated Data Augmentation

A Unified Label-Aware Contrastive Learning Framework for Few-Shot Named Entity Recognition

A Framework of Data Augmentation While Active Learning for Chinese Named Entity Recognition

Enhancing Cross-Lingual Few-Shot Named Entity Recognition by Prompt-Guiding

Knowledge-Enriched Prompt for Low-Resource Named Entity Recognition