Previous literatures show that pre-trained masked language models (MLMs) such as BERT can achieve competitive factual knowledge extraction performance on some datasets, indicating that MLMs can potentially be a reliable knowledge source. In this paper, we conduct a rigorous study to explore the underlying predicting mechanisms of MLMs over different extraction paradigms. By investigating the behaviors of MLMs, we find that previous decent performance mainly owes to the biased prompts which overfit dataset artifacts. Furthermore, incorporating illustrative cases and external contexts improve knowledge prediction mainly due to entity type guidance and golden answer leakage. Our findings shed light on the underlying predicting mechanisms of MLMs, and strongly question the previous conclusion that current MLMs can potentially serve as reliable factual knowledge bases.
@article{arxiv.2106.09231,
title = {Knowledgeable or Educated Guess? Revisiting Language Models as Knowledge Bases},
author = {Boxi Cao and Hongyu Lin and Xianpei Han and Le Sun and Lingyong Yan and Meng Liao and Tong Xue and Jin Xu},
journal= {arXiv preprint arXiv:2106.09231},
year = {2021}
}