Rapidly Bootstrapping a Question Answering Dataset for COVID-19

TL;DR

CovidQA is an early, manually created question answering dataset for COVID-19, designed to evaluate zero-shot and transfer learning models in a rapidly evolving pandemic context.

Contribution

It introduces the first publicly available COVID-19 QA dataset, providing a resource for evaluating models' zero-shot and transfer capabilities.

Findings

Transformer models outperform term-based techniques

Baseline models show limited performance, highlighting the challenge of COVID-19 QA

Dataset serves as a useful benchmark for COVID-19 related question answering

Abstract

We present CovidQA, the beginnings of a question answering dataset specifically designed for COVID-19, built by hand from knowledge gathered from Kaggle's COVID-19 Open Research Dataset Challenge. To our knowledge, this is the first publicly available resource of its type, and intended as a stopgap measure for guiding research until more substantial evaluation resources become available. While this dataset, comprising 124 question-article pairs as of the present version 0.1 release, does not have sufficient examples for supervised machine learning, we believe that it can be helpful for evaluating the zero-shot or transfer capabilities of existing models on topics specifically related to COVID-19. This paper describes our methodology for constructing the dataset and presents the effectiveness of a number of baselines, including term-based techniques and various transformer-based models. The dataset is available at http://covidqa.ai/