Knowing False Negatives: An Adversarial Training Method for Distantly Supervised Relation Extraction

Kailong Hao, Botao Yu, Wei Hu


Abstract
Distantly supervised relation extraction (RE) automatically aligns unstructured text with relation instances in a knowledge base (KB). Due to the incompleteness of current KBs, sentences implying certain relations may be annotated as N/A instances, which causes the so-called false negative (FN) problem. Current RE methods usually overlook this problem, inducing improper biases in both training and testing procedures. To address this issue, we propose a two-stage approach. First, it finds out possible FN samples by heuristically leveraging the memory mechanism of deep neural networks. Then, it aligns those unlabeled data with the training data into a unified feature space by adversarial training to assign pseudo labels and further utilize the information contained in them. Experiments on two wildly-used benchmark datasets demonstrate the effectiveness of our approach.
Anthology ID:
2021.emnlp-main.761
Volume:
Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
Month:
November
Year:
2021
Address:
Online and Punta Cana, Dominican Republic
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
9661–9672
Language:
URL:
https://aclanthology.org/2021.emnlp-main.761
DOI:
10.18653/v1/2021.emnlp-main.761
Bibkey:
Cite (ACL):
Kailong Hao, Botao Yu, and Wei Hu. 2021. Knowing False Negatives: An Adversarial Training Method for Distantly Supervised Relation Extraction. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 9661–9672, Online and Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Knowing False Negatives: An Adversarial Training Method for Distantly Supervised Relation Extraction (Hao et al., EMNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.emnlp-main.761.pdf
Video:
 https://aclanthology.org/2021.emnlp-main.761.mp4
Code
 nju-websoft/fan