MLMLM: Link Prediction with Mean Likelihood Masked Language Model
This addresses the problem of integrating scalable and interpretable knowledge representation for AI researchers and practitioners, though it is incremental as it builds on existing MLM methods.
The authors tackled the scalability issues of Knowledge Bases and interpretability issues of Masked Language Models by proposing MLMLM for link prediction, achieving State of the Art results on WN18RR and competitive results on FB15k-237, with convincing performance on unseen entities.
Knowledge Bases (KBs) are easy to query, verifiable, and interpretable. They however scale with man-hours and high-quality data. Masked Language Models (MLMs), such as BERT, scale with computing power as well as unstructured raw text data. The knowledge contained within those models is however not directly interpretable. We propose to perform link prediction with MLMs to address both the KBs scalability issues and the MLMs interpretability issues. To do that we introduce MLMLM, Mean Likelihood Masked Language Model, an approach comparing the mean likelihood of generating the different entities to perform link prediction in a tractable manner. We obtain State of the Art (SotA) results on the WN18RR dataset and the best non-entity-embedding based results on the FB15k-237 dataset. We also obtain convincing results on link prediction on previously unseen entities, making MLMLM a suitable approach to introducing new entities to a KB.