Abstract
We apply BERT to coreference resolution, achieving a new state of the art on the GAP (+11.5 F1) and OntoNotes (+3.9 F1) benchmarks. A qualitative analysis of model predictions indicates that, compared to ELMo and BERT-base, BERT-large is particularly better at distinguishing between related but distinct entities (e.g., President and CEO), but that there is still room for improvement in modeling document-level context, conversations, and mention paraphrasing. We will release all code and trained models upon publication.- Anthology ID:
- D19-1588
- Volume:
- Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
- Month:
- November
- Year:
- 2019
- Address:
- Hong Kong, China
- Editors:
- Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
- Venues:
- EMNLP | IJCNLP
- SIG:
- SIGDAT
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 5803–5808
- Language:
- URL:
- https://aclanthology.org/D19-1588
- DOI:
- 10.18653/v1/D19-1588
- Cite (ACL):
- Mandar Joshi, Omer Levy, Luke Zettlemoyer, and Daniel Weld. 2019. BERT for Coreference Resolution: Baselines and Analysis. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 5803–5808, Hong Kong, China. Association for Computational Linguistics.
- Cite (Informal):
- BERT for Coreference Resolution: Baselines and Analysis (Joshi et al., EMNLP-IJCNLP 2019)
- PDF:
- https://preview.aclanthology.org/ml4al-ingestion/D19-1588.pdf
- Code
- mandarjoshi90/coref + additional community code
- Data
- CoNLL, CoNLL-2012, GAP Coreference Dataset, OntoNotes 5.0