Reconciliation of Pre-trained Models and Prototypical Neural Networks in Few-shot Named Entity Recognition

TL;DR

This paper introduces a normalization technique to align pre-trained model embeddings with prototypical neural networks in few-shot NER, improving performance and offering insights into model synergy.

Contribution

A simple normalization method is proposed to mitigate embedding bias, enhancing the integration of pre-trained models with prototypical networks in few-shot NER.

Findings

Outperforms existing methods on nine benchmark datasets.

Achieves results comparable to state-of-the-art approaches.

Provides an analytical framework for addressing embedding biases.

Abstract

Incorporating large-scale pre-trained models with the prototypical neural networks is a de-facto paradigm in few-shot named entity recognition. Existing methods, unfortunately, are not aware of the fact that embeddings from pre-trained models contain a prominently large amount of information regarding word frequencies, biasing prototypical neural networks against learning word entities. This discrepancy constrains the two models' synergy. Thus, we propose a one-line-code normalization method to reconcile such a mismatch with empirical and theoretical grounds. Our experiments based on nine benchmark datasets show the superiority of our method over the counterpart models and are comparable to the state-of-the-art methods. In addition to the model enhancement, our work also provides an analytical viewpoint for addressing the general problems in few-shot name entity recognition or other tasks that rely on pre-trained models or prototypical neural networks.