Multilinguals at SemEval-2022 Task 11: Transformer Based Architecture for Complex NER

TL;DR

This paper explores the use of transformer-based models, specifically BERT, for complex named entity recognition in English, achieving significant improvements over baseline methods.

Contribution

It introduces a transformer-based architecture for complex NER and demonstrates its effectiveness through qualitative analysis and performance gains.

Findings

Best model beats baseline F1-score by over 9%

All models outperform baseline significantly

Qualitative analysis of multiple architectures

Abstract

We investigate the task of complex NER for the English language. The task is non-trivial due to the semantic ambiguity of the textual structure and the rarity of occurrence of such entities in the prevalent literature. Using pre-trained language models such as BERT, we obtain a competitive performance on this task. We qualitatively analyze the performance of multiple architectures for this task. All our models are able to outperform the baseline by a significant margin. Our best performing model beats the baseline F1-score by over 9%.