%0 Conference Proceedings %T How Pre-trained Language Models Capture Factual Knowledge? A Causal-Inspired Analysis %A Li, Shaobo %A Li, Xiaoguang %A Shang, Lifeng %A Dong, Zhenhua %A Sun, Chengjie %A Liu, Bingquan %A Ji, Zhenzhou %A Jiang, Xin %A Liu, Qun %Y Muresan, Smaranda %Y Nakov, Preslav %Y Villavicencio, Aline %S Findings of the Association for Computational Linguistics: ACL 2022 %D 2022 %8 May %I Association for Computational Linguistics %C Dublin, Ireland %F li-etal-2022-pre %X Recently, there has been a trend to investigate the factual knowledge captured by Pre-trained Language Models (PLMs). Many works show the PLMs’ ability to fill in the missing factual words in cloze-style prompts such as ”Dante was born in [MASK].” However, it is still a mystery how PLMs generate the results correctly: relying on effective clues or shortcut patterns? We try to answer this question by a causal-inspired analysis that quantitatively measures and evaluates the word-level patterns that PLMs depend on to generate the missing words. We check the words that have three typical associations with the missing words: knowledge-dependent, positionally close, and highly co-occurred. Our analysis shows: (1) PLMs generate the missing factual words more by the positionally close and highly co-occurred words than the knowledge-dependent words; (2) the dependence on the knowledge-dependent words is more effective than the positionally close and highly co-occurred words. Accordingly, we conclude that the PLMs capture the factual knowledge ineffectively because of depending on the inadequate associations. %R 10.18653/v1/2022.findings-acl.136 %U https://aclanthology.org/2022.findings-acl.136 %U https://doi.org/10.18653/v1/2022.findings-acl.136 %P 1720-1732