Ensemble Transformer for Efficient and Accurate Ranking Tasks: an Application to Question Answering Systems
This work addresses efficiency and accuracy challenges in question answering systems, offering a practical solution for real-world applications where computational resources are limited, though it is incremental as it builds on existing distillation techniques.
The paper tackles the problem of high computational costs in large transformer models for Answer Sentence Selection (AS2) by proposing CERBERUS, an efficient neural network that distills an ensemble of large transformers into a smaller model, outperforming single-model distillations and rivaling state-of-the-art models with 2.7x more parameters and 2.5x slower runtime.
Large transformer models can highly improve Answer Sentence Selection (AS2) tasks, but their high computational costs prevent their use in many real-world applications. In this paper, we explore the following research question: How can we make the AS2 models more accurate without significantly increasing their model complexity? To address the question, we propose a Multiple Heads Student architecture (named CERBERUS), an efficient neural network designed to distill an ensemble of large transformers into a single smaller model. CERBERUS consists of two components: a stack of transformer layers that is used to encode inputs, and a set of ranking heads; unlike traditional distillation technique, each of them is trained by distilling a different large transformer architecture in a way that preserves the diversity of the ensemble members. The resulting model captures the knowledge of heterogeneous transformer models by using just a few extra parameters. We show the effectiveness of CERBERUS on three English datasets for AS2; our proposed approach outperforms all single-model distillations we consider, rivaling the state-of-the-art large AS2 models that have 2.7x more parameters and run 2.5x slower. Code for our model is available at https://github.com/amazon-research/wqa-cerberus