Improving Downstream Task Performance by Treating Numbers as Entities

TL;DR

This paper proposes classifying numbers as entities in NLP models to enhance performance on tasks like fill-in-the-blank and question answering, leveraging the models' inherent numeracy.

Contribution

It introduces a novel classification of numbers as entities, improving NLP task performance beyond standard BERT and RoBERTa baselines.

Findings

Improved accuracy on fill-in-the-blank tasks

Enhanced question answering performance

Outperforms baseline models

Abstract

Numbers are essential components of text, like any other word tokens, from which natural language processing (NLP) models are built and deployed. Though numbers are typically not accounted for distinctly in most NLP tasks, there is still an underlying amount of numeracy already exhibited by NLP models. In this work, we attempt to tap this potential of state-of-the-art NLP models and transfer their ability to boost performance in related tasks. Our proposed classification of numbers into entities helps NLP models perform well on several tasks, including a handcrafted Fill-In-The-Blank (FITB) task and on question answering using joint embeddings, outperforming the BERT and RoBERTa baseline classification.