Class-Adaptive Self-Training for Relation Extraction with Incompletely Annotated Training Data

Qingyu Tan, Lu Xu, Lidong Bing, Hwee Tou Ng


Abstract
Relation extraction (RE) aims to extract relations from sentences and documents. Existing relation extraction models typically rely on supervised machine learning. However, recent studies showed that many RE datasets are incompletely annotated. This is known as the false negative problem in which valid relations are falsely annotated as ‘no_relation’. Models trained with such data inevitably make similar mistakes during the inference stage. Self-training has been proven effective in alleviating the false negative problem. However, traditional self-training is vulnerable to confirmation bias and exhibits poor performance in minority classes. To overcome this limitation, we proposed a novel class-adaptive re-sampling self-training framework. Specifically, we re-sampled the pseudo-labels for each class by precision and recall scores. Our re-sampling strategy favored the pseudo-labels of classes with high precision and low recall, which improved the overall recall without significantly compromising precision. We conducted experiments on document-level and biomedical relation extraction datasets, and the results showed that our proposed self-training framework consistently outperforms existing competitive methods on the Re-DocRED and ChemDisgene datasets when the training data are incompletely annotated.
Anthology ID:
2023.findings-acl.549
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8630–8643
Language:
URL:
https://aclanthology.org/2023.findings-acl.549
DOI:
10.18653/v1/2023.findings-acl.549
Bibkey:
Cite (ACL):
Qingyu Tan, Lu Xu, Lidong Bing, and Hwee Tou Ng. 2023. Class-Adaptive Self-Training for Relation Extraction with Incompletely Annotated Training Data. In Findings of the Association for Computational Linguistics: ACL 2023, pages 8630–8643, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Class-Adaptive Self-Training for Relation Extraction with Incompletely Annotated Training Data (Tan et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.549.pdf
Video:
 https://aclanthology.org/2023.findings-acl.549.mp4