Contemporary question answering (QA) systems, including transformer-based architectures, suffer from increasing computational and model complexity which render them inefficient for real-world applications with limited resources. Further, training or even fine-tuning such models requires a vast amount of labeled data which is often not available for the task at hand. In this manuscript, we conduct a comprehensive analysis of the mentioned challenges and introduce suitable countermeasures. We propose a novel knowledge distillation (KD) approach to reduce the parameter and model complexity of a pre-trained BERT system and utilize multiple active learning (AL) strategies for immense reduction in annotation efforts. In particular, we demonstrate that our model achieves the performance of a 6-layer TinyBERT and DistilBERT, whilst using only 2% of their total parameters. Finally, by the integration of our AL approaches into the BERT framework, we show that state-of-the-art results on the SQuAD dataset can be achieved when we only use 20% of the training data.
@article{arxiv.2109.12662,
title = {Improving Question Answering Performance Using Knowledge Distillation and Active Learning},
author = {Yasaman Boreshban and Seyed Morteza Mirbostani and Gholamreza Ghassem-Sani and Seyed Abolghasem Mirroshandel and Shahin Amiriparian},
journal= {arXiv preprint arXiv:2109.12662},
year = {2021}
}