Jinhyuk Lee, Jaewoo Kang, Miyoung Ko, Hyunjae Kim, Gangwoo Kim
We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
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.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2004.14602")
get_code_for_paper("2004.14602")
have("2004.14602")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2004.14602.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2004.14602)
Connect an agent — have() is free.