Few-Shot Named Entity Recognition: A Comprehensive Study
By Jiaxin Huang, Chunyuan Li, Krishan Subudhi, Damien Jose, Shobana Balakrishnan, Weizhu Chen, Baolin Peng, Jianfeng Gao, Jiawei Han. 2020
83 citations recorded by the source corpus.
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.
Subfields: Artificial Intelligence
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Zhiyuan Liu
- Luke S. Zettlemoyer
- Mohit Bansal
- Yue Zhang
- Michael Lewis
- Percy Liang
- Maosong Sun
- Jianfeng Gao
Citations in the corpus
- Few-Shot Class-Incremental Learning for Named Entity Recognition (2022, background): “Al-ternatively, Huang et al. (2020) avoids overfitting of few-shot learning by augmenting with noisy or unlabeled data from the web.”
- Few-NERD: A Few-shot Named Entity Recognition Dataset (2021, background): “Few-shot NER is a considerably challenging and practical problem that could facilitate the understanding of textual knowledge for neural model (Huang et al., 2020).”
- Template-Based Named Entity Recognition Using BART (2021, methodology): “For example, the performance of Huang et al. (2020) increases only 1.9% F1 score when the number of instances per entity type increase from 10 to 500.”
- Question Answering Infused Pre-training of General-Purpose Contextualized Representations (2021, methodology): “We use the few-shot splits from Huang et al. (2020) that include 5 examples per entity type.”
- Large language models for generative information extraction: a survey (2023, unknown): “Di ff erent IE tasks vary a lot, with di ff erent optimization objectives and task-specific schemas, requiring separate models to handle the complexity of di ff erent IE tasks, settings, and scenarios [4].”