How Context Affects Language Models' Factual Predictions
doi:10.24432/C5201W
TL;DR
When pre-trained on large unsupervised textual corpora, language models are able to store and retrieve factual knowledge to some extent, making it possible to use them directly for zero-shot cloze-style question answering. However, storing factual knowledge in a fixed number of weights of a language model clearly has limitations. Previous approaches have successfully provided access to information outside the model weights using supervised architectures that combine an information retrieval system with a machine reading component. In this paper, we go one step further and integrate information from a retrieval system with a pre-trained language model in a purely unsupervised way. We report that augmenting pre-trained language models in this way dramatically improves performance and that it is competitive with a supervised machine reading baseline without requiring any supervised training. Furthermore, processing query and context with different segment tokens allows BERT to utilize its Next Sentence Prediction pre-trained classifier to determine whether the context is relevant or not, substantially improving BERT's zero-shot cloze-style question-answering performance and making its predictions robust to noisy contexts.
Citation
@inproceedings{
petroni2020how,
title={How Context Affects Language Models' Factual Predictions},
author={Fabio Petroni and Patrick Lewis and Aleksandra Piktus and Tim Rockt{\"a}schel and Yuxiang Wu and Alexander H. Miller and Sebastian Riedel},
booktitle={Automated Knowledge Base Construction},
year={2020},
url={https://openreview.net/forum?id=025X0zPfn},
doi={10.24432/C5201W}
}