Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling
AbstractDocument-level relation extraction (RE) poses new challenges compared to its sentence-level counterpart. One document commonly contains multiple entity pairs, and one entity pair occurs multiple times in the document associated with multiple possible relations. In this paper, we propose two novel techniques, adaptive thresholding and localized context pooling, to solve the multi-label and multi-entity problems. The adaptive thresholding replaces the global threshold for multi-label classification in the prior work with a learnable entities-dependent threshold. The localized context pooling directly transfers attention from pre-trained language models to locate relevant context that is useful to decide the relation. We experiment on three document-level RE benchmark datasets: DocRED, a recently released large-scale RE dataset, and two datasets CDRand GDA in the biomedical domain. Our ATLOP (Adaptive Thresholding and Localized cOntext Pooling) model achieves an F1 score of 63.4, and also significantly outperforms existing models on both CDR and GDA. We have released our code at https://github.com/wzhouad/ATLOP.
How to Cite
Zhou, W., Huang, K., Ma, T., & Huang, J. (2021). Document-Level Relation Extraction with Adaptive Thresholding and Localized Context Pooling. Proceedings of the AAAI Conference on Artificial Intelligence, 35(16), 14612-14620. Retrieved from https://ojs.aaai.org/index.php/AAAI/article/view/17717
AAAI Technical Track on Speech and Natural Language Processing III