UNIFIEDQA: Crossing Format Boundaries with a Single QA System

Daniel Khashabi, Sewon Min, Tushar Khot, Ashish Sabharwal, Oyvind Tafjord, Peter Clark, Hannaneh Hajishirzi


Abstract
Question answering (QA) tasks have been posed using a variety of formats, such as extractive span selection, multiple choice, etc. This has led to format-specialized models, and even to an implicit division in the QA community. We argue that such boundaries are artificial and perhaps unnecessary, given the reasoning abilities we seek to teach are not governed by the format. As evidence, we use the latest advances in language modeling to build a single pre-trained QA model, UNIFIEDQA, that performs well across 19 QA datasets spanning 4 diverse formats. UNIFIEDQA performs on par with 8 different models that were trained on individual datasets themselves. Even when faced with 12 unseen datasets of observed formats, UNIFIEDQA performs surprisingly well, showing strong generalization from its outof-format training data. Finally, simply finetuning this pre trained QA model into specialized models results in a new state of the art on 10 factoid and commonsense question answering datasets, establishing UNIFIEDQA as a strong starting point for building QA systems.
Anthology ID:
2020.findings-emnlp.171
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Venues:
EMNLP | Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1896–1907
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.171
DOI:
10.18653/v1/2020.findings-emnlp.171
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.findings-emnlp.171.pdf
Optional supplementary material:
 2020.findings-emnlp.171.OptionalSupplementaryMaterial.pdf