Inducing Relational Knowledge from BERT


  • Zied Bouraoui CRIL
  • Jose Camacho-Collados Cardiff University
  • Steven Schockaert Cardiff University



One of the most remarkable properties of word embeddings is the fact that they capture certain types of semantic and syntactic relationships. Recently, pre-trained language models such as BERT have achieved groundbreaking results across a wide range of Natural Language Processing tasks. However, it is unclear to what extent such models capture relational knowledge beyond what is already captured by standard word embeddings. To explore this question, we propose a methodology for distilling relational knowledge from a pre-trained language model. Starting from a few seed instances of a given relation, we first use a large text corpus to find sentences that are likely to express this relation. We then use a subset of these extracted sentences as templates. Finally, we fine-tune a language model to predict whether a given word pair is likely to be an instance of some relation, when given an instantiated template for that relation as input.




How to Cite

Bouraoui, Z., Camacho-Collados, J., & Schockaert, S. (2020). Inducing Relational Knowledge from BERT. Proceedings of the AAAI Conference on Artificial Intelligence, 34(05), 7456-7463.



AAAI Technical Track: Natural Language Processing