End-to-end Neural Coreference Resolution

Kenton Lee, Luheng He, Mike Lewis, Luke Zettlemoyer


Abstract
We introduce the first end-to-end coreference resolution model and show that it significantly outperforms all previous work without using a syntactic parser or hand-engineered mention detector. The key idea is to directly consider all spans in a document as potential mentions and learn distributions over possible antecedents for each. The model computes span embeddings that combine context-dependent boundary representations with a head-finding attention mechanism. It is trained to maximize the marginal likelihood of gold antecedent spans from coreference clusters and is factored to enable aggressive pruning of potential mentions. Experiments demonstrate state-of-the-art performance, with a gain of 1.5 F1 on the OntoNotes benchmark and by 3.1 F1 using a 5-model ensemble, despite the fact that this is the first approach to be successfully trained with no external resources.
Anthology ID:
D17-1018
Volume:
Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing
Month:
September
Year:
2017
Address:
Copenhagen, Denmark
Editors:
Martha Palmer, Rebecca Hwa, Sebastian Riedel
Venue:
EMNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
188–197
Language:
URL:
https://aclanthology.org/D17-1018
DOI:
10.18653/v1/D17-1018
Bibkey:
Cite (ACL):
Kenton Lee, Luheng He, Mike Lewis, and Luke Zettlemoyer. 2017. End-to-end Neural Coreference Resolution. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 188–197, Copenhagen, Denmark. Association for Computational Linguistics.
Cite (Informal):
End-to-end Neural Coreference Resolution (Lee et al., EMNLP 2017)
Copy Citation:
PDF:
https://aclanthology.org/D17-1018.pdf
Video:
 https://aclanthology.org/D17-1018.mp4
Code
 kentonl/e2e-coref +  additional community code
Data
CoNLLCoNLL-2012OntoNotes 5.0