You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Source code of the proposed models and experiments in the paper "[OKGIT: Open Knowledge Graph Link Prediction with Implicit Types]()" to be presented in the Findings of ACL 2021.
The code is based on CaRE and LAMA. The datasets used in this work can be found here. The pre-trained language models (BERT-base, BERT-large, and RoBERTa) used in the paper can be found here.
The OKGIT model is described in the following paper:
@inproceedings{chandrahas-talukdar-2021-okgit,
title = "{OKGIT}: {O}pen Knowledge Graph Link Prediction with Implicit Types",
author = "Chandrahas, . and Talukdar, Partha",
booktitle = "Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.findings-acl.225",
doi = "10.18653/v1/2021.findings-acl.225",
pages = "2546--2559",
}
About
Source code of the proposed models and experiments in the paper "[OKGIT: Open Knowledge Graph Link Prediction with Implicit Types]()" to be presented in the Findings of ACL 2021.