Abstract
In relation extraction with distant supervision, noisy labels make it difficult to train quality models. Previous neural models addressed this problem using an attention mechanism that attends to sentences that are likely to express the relations. We improve such models by combining the distant supervision data with an additional directly-supervised data, which we use as supervision for the attention weights. We find that joint training on both types of supervision leads to a better model because it improves the model’s ability to identify noisy sentences. In addition, we find that sigmoidal attention weights with max pooling achieves better performance over the commonly used weighted average attention in this setup. Our proposed method achieves a new state-of-the-art result on the widely used FB-NYT dataset.- Anthology ID:
- N19-1184
- Volume:
- Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers)
- Month:
- June
- Year:
- 2019
- Address:
- Minneapolis, Minnesota
- Editors:
- Jill Burstein, Christy Doran, Thamar Solorio
- Venue:
- NAACL
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 1858–1867
- Language:
- URL:
- https://aclanthology.org/N19-1184
- DOI:
- 10.18653/v1/N19-1184
- Cite (ACL):
- Iz Beltagy, Kyle Lo, and Waleed Ammar. 2019. Combining Distant and Direct Supervision for Neural Relation Extraction. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 1858–1867, Minneapolis, Minnesota. Association for Computational Linguistics.
- Cite (Informal):
- Combining Distant and Direct Supervision for Neural Relation Extraction (Beltagy et al., NAACL 2019)
- PDF:
- https://preview.aclanthology.org/fix-dup-bibkey/N19-1184.pdf
- Code
- allenai/comb_dist_direct_relex