Global Pointer: Novel Efficient Span-based Approach for Named Entity Recognition

TL;DR

The paper introduces Global Pointer, a novel span-based NER framework that efficiently captures semantic information using a global view and attention mechanisms, outperforming existing methods on benchmark datasets.

Contribution

It proposes a new span-based NER model with a global attention mechanism, a novel loss function, and an efficient training method, improving accuracy and computational efficiency.

Findings

Outperforms existing span-based NER methods on benchmarks.

The new loss function effectively addresses label imbalance.

The model demonstrates improved accuracy and efficiency.

Abstract

Named entity recognition (NER) task aims at identifying entities from a piece of text that belong to predefined semantic types such as person, location, organization, etc. The state-of-the-art solutions for flat entities NER commonly suffer from capturing the fine-grained semantic information in underlying texts. The existing span-based approaches overcome this limitation, but the computation time is still a concern. In this work, we propose a novel span-based NER framework, namely Global Pointer (GP), that leverages the relative positions through a multiplicative attention mechanism. The ultimate goal is to enable a global view that considers the beginning and the end positions to predict the entity. To this end, we design two modules to identify the head and the tail of a given entity to enable the inconsistency between the training and inference processes. Moreover, we introduce a novel classification loss function to address the imbalance label problem. In terms of parameters, we introduce a simple but effective approximate method to reduce the training parameters. We extensively evaluate GP on various benchmark datasets. Our extensive experiments demonstrate that GP can outperform the existing solution. Moreover, the experimental results show the efficacy of the introduced loss function compared to softmax and entropy alternatives.