NeurIPS 2020 EfficientQA Competition: Systems, Analyses and Lessons Learned
This paper is significant for the open-domain QA research community by analyzing the results and lessons learned from a competition focused on efficiency constraints.
This paper reviews the NeurIPS 2020 EfficientQA competition, which challenged participants to develop open-domain question answering systems that could provide correct answers while adhering to strict on-disk memory budgets. The competition aimed to explore the trade-off between storing retrieval corpora and learned model parameters.
We review the EfficientQA competition from NeurIPS 2020. The competition focused on open-domain question answering (QA), where systems take natural language questions as input and return natural language answers. The aim of the competition was to build systems that can predict correct answers while also satisfying strict on-disk memory budgets. These memory budgets were designed to encourage contestants to explore the trade-off between storing retrieval corpora or the parameters of learned models. In this report, we describe the motivation and organization of the competition, review the best submissions, and analyze system predictions to inform a discussion of evaluation for open-domain QA.