Anatomy of OntoGUMAdapting GUM to the OntoNotes Scheme to Evaluate Robustness of SOTA Coreference Algorithms

Yilun Zhu, Sameer Pradhan, Amir Zeldes


Abstract
SOTA coreference resolution produces increasingly impressive scores on the OntoNotes benchmark. However lack of comparable data following the same scheme for more genres makes it difficult to evaluate generalizability to open domain data. Zhu et al. (2021) introduced the creation of the OntoGUM corpus for evaluating geralizability of the latest neural LM-based end-to-end systems. This paper covers details of the mapping process which is a set of deterministic rules applied to the rich syntactic and discourse annotations manually annotated in the GUM corpus. Out-of-domain evaluation across 12 genres shows nearly 15-20% degradation for both deterministic and deep learning systems, indicating a lack of generalizability or covert overfitting in existing coreference resolution models.
Anthology ID:
2021.crac-1.15
Volume:
Proceedings of the Fourth Workshop on Computational Models of Reference, Anaphora and Coreference
Month:
November
Year:
2021
Address:
Punta Cana, Dominican Republic
Venues:
CRAC | EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
141–149
Language:
URL:
https://aclanthology.org/2021.crac-1.15
DOI:
10.18653/v1/2021.crac-1.15
Bibkey:
Cite (ACL):
Yilun Zhu, Sameer Pradhan, and Amir Zeldes. 2021. Anatomy of OntoGUM—Adapting GUM to the OntoNotes Scheme to Evaluate Robustness of SOTA Coreference Algorithms. In Proceedings of the Fourth Workshop on Computational Models of Reference, Anaphora and Coreference, pages 141–149, Punta Cana, Dominican Republic. Association for Computational Linguistics.
Cite (Informal):
Anatomy of OntoGUM—Adapting GUM to the OntoNotes Scheme to Evaluate Robustness of SOTA Coreference Algorithms (Zhu et al., CRAC 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.crac-1.15.pdf
Data
GAP Coreference DatasetOntoGUMWikiCoref