Louis Clouatre, Philippe Trempe, Amal Zouaq et Sarath Chandar Anbil Parthipan
Communication écrite (2021)
|
Libre accès au plein texte de ce document Version officielle de l'éditeur Conditions d'utilisation: Creative Commons: Attribution (CC BY) Télécharger (474kB) |
Abstract
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 these 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. By committing the knowledge embedded in MLMs to a KB, it becomes interpretable. 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 SotA results on the Precision@1 metric on the WikidataM5 inductive and transductive setting. We also obtain convincing results on link prediction on previously unseen entities, making MLMLM a suitable approach to introducing new entities to a KB.
Sujet(s): |
2700 Technologie de l'information > 2705 Logiciels et développement 2700 Technologie de l'information > 2706 Génie logiciel |
---|---|
Département: | Département de génie informatique et génie logiciel |
Organismes subventionnaires: | Apogée Canada, Canada First Research Excellence Fund program, École Polytechnique Startup Fund PIED, Canada CIFAR AI Chair, GRSNG / NSERC - Discovery Grant |
URL de PolyPublie: | https://publications.polymtl.ca/10613/ |
Nom de la conférence: | The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021) |
Lieu de la conférence: | Bangkok, Thailand |
Date(s) de la conférence: | 2021-08-01 - 2021-08-06 |
Maison d'édition: | Association for Computational Linguistics |
DOI: | 10.18653/v1/2021.findings-acl.378 |
URL officielle: | https://doi.org/10.18653/v1/2021.findings-acl.378 |
Date du dépôt: | 19 déc. 2022 13:47 |
Dernière modification: | 27 sept. 2024 00:14 |
Citer en APA 7: | Clouatre, L., Trempe, P., Zouaq, A., & Anbil Parthipan, S. C. (août 2021). MLMLM: link prediction with mean likelihood masked language model [Communication écrite]. The Joint Conference of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (ACL-IJCNLP 2021), Bangkok, Thailand. https://doi.org/10.18653/v1/2021.findings-acl.378 |
---|---|
Statistiques
Total des téléchargements à partir de PolyPublie
Téléchargements par année

Provenance des téléchargements

Dimensions