Abstract
Many extractive question answering models are trained to predict start and end positions of answers. The choice of predicting answers as positions is mainly due to its simplicity and effectiveness. In this study, we hypothesize that when the distribution of the answer positions is highly skewed in the training set (e.g., answers lie only in the k-th sentence of each passage), QA models predicting answers as positions can learn spurious positional cues and fail to give answers in different positions. We first illustrate this position bias in popular extractive QA models such as BiDAF and BERT and thoroughly examine how position bias propagates through each layer of BERT. To safely deliver position information without position bias, we train models with various de-biasing methods including entropy regularization and bias ensembling. Among them, we found that using the prior distribution of answer positions as a bias model is very effective at reducing position bias, recovering the performance of BERT from 37.48% to 81.64% when trained on a biased SQuAD dataset.
Original language | English |
---|---|
Title of host publication | EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference |
Publisher | Association for Computational Linguistics (ACL) |
Pages | 1109-1121 |
Number of pages | 13 |
ISBN (Electronic) | 9781952148606 |
Publication status | Published - 2020 |
Event | 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020 - Virtual, Online Duration: 2020 Nov 16 → 2020 Nov 20 |
Publication series
Name | EMNLP 2020 - 2020 Conference on Empirical Methods in Natural Language Processing, Proceedings of the Conference |
---|
Conference
Conference | 2020 Conference on Empirical Methods in Natural Language Processing, EMNLP 2020 |
---|---|
City | Virtual, Online |
Period | 20/11/16 → 20/11/20 |
Bibliographical note
Publisher Copyright:© 2020 Association for Computational Linguistics
ASJC Scopus subject areas
- Information Systems
- Computer Science Applications
- Computational Theory and Mathematics