Few-Shot Named Entity Recognition: A Comprehensive Study

Abstract

This paper presents a comprehensive study to efficiently build named entity recognition (NER) systems when a small number of in-domain labeled data is available. Based upon recent Transformer-based self-supervised pre-trained language models (PLMs), we investigate three orthogonal schemes to improve the model generalization ability for few-shot settings: (1) meta-learning to construct prototypes for different entity types, (2) supervised pre-training on noisy web data to extract entity-related generic representations and (3) self-training to leverage unlabeled in-domain data. Different combinations of these schemes are also considered. We perform extensive empirical comparisons on 10 public NER datasets with various proportions of labeled data, suggesting useful insights for future research. Our experiments show that (i) in the few-shot learning setting, the proposed NER schemes significantly improve or outperform the commonly used baseline, a PLM-based linear classifier fine-tuned on domain labels; (ii) We create new state-of-the-art results on both few-shot and training-free settings compared with existing methods. We will release our code and pre-trained models for reproducible research.

References in corpus (15)

• Prototypical Networks for Few-shot Learning
• How transferable are features in deep neural networks?
• CTRL: A Conditional Transformer Language Model for Controllable Generation
• Rethinking Pre-training and Self-training
• Snips Voice Platform: an embedded Spoken Language Understanding system for private-by-design voice interfaces
• Sentence Encoders on STILTs: Supplementary Training on Intermediate Labeled-data Tasks
• Few-shot classification in Named Entity Recognition Task
• SOLOIST: Building Task Bots at Scale with Transfer Learning and Machine Teaching
• Self-training Improves Pre-training for Natural Language Understanding
• Uncertainty-aware Self-training for Text Classification with Few Labels
• Example-Based Named Entity Recognition
• Self-supervised Pre-training with Hard Examples Improves Visual Representations
• Robust Conversational AI with Grounded Text Generation
• Repulsive Attention: Rethinking Multi-head Attention as Bayesian Inference
• ReMP: Rectified Metric Propagation for Few-Shot Learning

Cited by in corpus (7)

• Cross-domain Named Entity Recognition via Graph Matching
• Recent Advances in Named Entity Recognition: A Comprehensive Survey and Comparative Study
• From Zero to Hero: Harnessing Transformers for Biomedical Named Entity Recognition in Zero- and Few-shot Contexts
• Template-Based Named Entity Recognition Using BART
• Few-NERD: A Few-Shot Named Entity Recognition Dataset
• Few-Sample Named Entity Recognition for Security Vulnerability Reports by Fine-Tuning Pre-Trained Language Models
• Learning from Language Description: Low-shot Named Entity Recognition via Decomposed Framework