Empower Distantly Supervised Relation Extraction with Collaborative Adversarial Training


  • Tao Chen Zhejiang University
  • Haochen Shi Zhejiang University
  • Liyuan Liu University of Illinois at Urbana Champaign
  • Siliang Tang Zhejiang University
  • Jian Shao Zhejiang University
  • Zhigang Chen iFLYTEK Research
  • Yueting Zhuang Zhejiang University




Information Extraction


With recent advances in distantly supervised (DS) relation extraction (RE), considerable attention is attracted to leverage multi-instance learning (MIL) to distill high-quality supervision from the noisy DS. Here, we go beyond label noise and identify the key bottleneck of DS-MIL to be its low data utilization: as high-quality supervision being refined by MIL, MIL abandons a large amount of training instances, which leads to a low data utilization and hinders model training from having abundant supervision. In this paper, we propose collaborative adversarial training to improve the data utilization, which coordinates virtual adversarial training (VAT) and adversarial training (AT) at different levels. Specifically, since VAT is label-free, we employ the instance-level VAT to recycle instances abandoned by MIL. Besides, we deploy AT at the bag-level to unleash the full potential of the high-quality supervision got by MIL. Our proposed method brings consistent improvements (∼ 5 absolute AUC score) to the previous state of the art, which verifies the importance of the data utilization issue and the effectiveness of our method.




How to Cite

Chen, T., Shi, H., Liu, L., Tang, S., Shao, J., Chen, Z., & Zhuang, Y. (2021). Empower Distantly Supervised Relation Extraction with Collaborative Adversarial Training. Proceedings of the AAAI Conference on Artificial Intelligence, 35(14), 12675-12682. https://doi.org/10.1609/aaai.v35i14.17501



AAAI Technical Track on Speech and Natural Language Processing I