Model Cascading: Towards Jointly Improving Efficiency and Accuracy of NLP Systems

TL;DR

This paper explores model cascading in NLP, combining models of different sizes to enhance efficiency and accuracy, demonstrating significant computational savings and accuracy improvements across various tasks.

Contribution

It introduces and empirically evaluates model cascading as a simple method to improve NLP system efficiency and accuracy using multiple models of varying capacities.

Findings

Cascading saves up to 88.93% computation cost.

Achieves up to 2.18% accuracy improvement.

Adding more models increases efficiency gains.

Abstract

Do all instances need inference through the big models for a correct prediction? Perhaps not; some instances are easy and can be answered correctly by even small capacity models. This provides opportunities for improving the computational efficiency of systems. In this work, we present an explorative study on 'model cascading', a simple technique that utilizes a collection of models of varying capacities to accurately yet efficiently output predictions. Through comprehensive experiments in multiple task settings that differ in the number of models available for cascading (K value), we show that cascading improves both the computational efficiency and the prediction accuracy. For instance, in K=3 setting, cascading saves up to 88.93% computation cost and consistently achieves superior prediction accuracy with an improvement of up to 2.18%. We also study the impact of introducing additional models in the cascade and show that it further increases the efficiency improvements. Finally, we hope that our work will facilitate development of efficient NLP systems making their widespread adoption in real-world applications possible.