--- title: "Using Type Information to Improve Entity — Khosla 2020" description: "Neural coreference resolution (CR) systems built on contextual encoders such as BERT usually ignore the entity-type labels that many CR corpora already…" authors: ["Sopan Khosla", "Carolyn Rose"] year: 2020 venue: "Proceedings of the First Workshop on Computational Approaches to Discourse" source: "https://aclanthology.org/2020.codi-1.3/" doi: "10.18653/v1/2020.codi-1.3" arxiv: "2010.05738" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/papers/khosla-2020-type-information-improve-entity-coreference/" --- [Elamur Research](https://md.research.elamur.ai/) · [this page in HTML](https://research.elamur.ai/free/papers/khosla-2020-type-information-improve-entity-coreference/) [home](https://md.research.elamur.ai/) / [papers](https://md.research.elamur.ai/free/papers/) / [2020](https://md.research.elamur.ai/free/papers/2020/page/5/) / Using Type Information to Improve Entity Coreference Resolution # Using Type Information to Improve Entity Coreference Resolution **Authors:** [Sopan Khosla](https://md.research.elamur.ai/free/authors/sopan-khosla/), [Carolyn Rose](https://md.research.elamur.ai/free/authors/carolyn-rose/) · **Year:** 2020 · **Venue:** Proceedings of the First Workshop on Computational Approaches to Discourse · **Licence:** [CC BY 4.0 (ACL Anthology)](https://creativecommons.org/licenses/by/4.0/) **Links:** [Publisher's page](https://aclanthology.org/2020.codi-1.3/) · [PDF](https://aclanthology.org/2020.codi-1.3.pdf) · [DOI 10.18653/v1/2020.codi-1.3](https://doi.org/10.18653/v1/2020.codi-1.3) · [arXiv 2010.05738](https://arxiv.org/abs/2010.05738) · [OpenAlex](https://openalex.org/W3093459127) **Citations:** 16 citations (openalex-snapshot · 2026-10-14) · 11 in our library, of the 140673 papers whose bibliographies we read ## Summary Neural coreference resolution (CR) systems built on contextual encoders such as BERT usually ignore the entity-type labels that many CR corpora already contain. Khosla and Rose test whether making these types explicit helps. They start from the mention-ranking model of Bamman et al. (2020), which derives from Lee et al. (2017), and add type knowledge in two places. First, each mention's span vector is extended with its type (+ET-self). Second, the pairwise scorer receives a binary flag marking whether two candidate mentions differ in type (+ET-cross). This flag acts as a soft filter rather than a hard one, so bridging-anaphora cases are not ruled out. Experiments use gold mentions on LitBank, EmailCoref, WikiCoref and OntoNotes. Each corpus is tested with its own tag set and with a shared coarse set. Both gold and automatically predicted types give modest but consistent gains over the baseline, and predicted types still beat it significantly. The study shows that explicit semantic types remain useful even for BERT-based CR. ## Key points - The baseline is the BERT-based model of Bamman et al. (2020), which handles only mention linking over gold mentions. The combined system, +ET, merges +ET-self and +ET-cross. - Four corpora are used: - LitBank: 100 literary texts, with 83.1% of mentions referring to persons. - EmailCoref: 46 threads containing 245 messages. - WikiCoref: 30 Wikipedia documents, split 24/3/3, with types taken from Freebase. - OntoNotes: a much larger type inventory. - Tag granularity varies widely across corpora, from 4 types in EmailCoref to 18 in OntoNotes. The authors therefore also map all corpora to a common set of PER, ORG, LOC, FAC and OTHER. - A separate type predictor uses the mention span and its sentence context. It struggles with demonstrative pronouns (*this*, *that*, *it*), because these carry little type signal. - An EmailCoref case study shows type-aware models fixing errors that the baseline makes. Predicted types still fail when sentence-level context is insufficient. *Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.* Fields: [Linguistics & languages](https://md.research.elamur.ai/free/shelves/linguistics/) Topics: [Coreference resolution](https://md.research.elamur.ai/free/topics/coreference-resolution/) ## Cite The paper itself, not this page. BibTeX from the library's record of the work. ```bibtex @inproceedings{khosla2020using, title = {{Using Type Information to Improve Entity Coreference Resolution}}, author = {Khosla, Sopan and Rose, Carolyn}, booktitle = {Proceedings of the First Workshop on Computational Approaches to Discourse}, year = {2020}, doi = {10.18653/v1/2020.codi-1.3}, eprint = {2010.05738}, archivePrefix = {arXiv}, url = {https://doi.org/10.18653/v1/2020.codi-1.3} } ``` - [← previous Unsupervised Translation of Programming Languages](https://md.research.elamur.ai/free/papers/lachaux-2020-unsupervised-translation-programming-languages/) - [next → USR: An Unsupervised and Reference Free Evaluation Metric for Dialog Generation](https://md.research.elamur.ai/free/papers/mehri-2020-usr-unsupervised-reference-free-evaluation/) - [↑ back to papers of 2020](https://md.research.elamur.ai/free/papers/2020/page/5/) ## References 4 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. David Bamman, Olivia Lewke, and Anya Mansoor. 2020. An annotated dataset of coreference in english literature. In Proceedings of The 12th Language Resources and Evaluation Conference, pages 44–54. [find](https://scholar.google.com/scholar?q=David+Bamman%2C+Olivia+Lewke%2C+and+Anya+Mansoor%2E+2020%2E+An+annotated+dataset+of+coreference+in+english+literature%2E+In+Proceedings+of+The+12th+Language+Resources+and+Evaluation+Conference%2C+pages+44%E2%80%9354%2E) [our card](https://md.research.elamur.ai/free/papers/bamman-2020-annotated-dataset-coreference-english-literature/) 2. Eric Bengtson and Dan Roth. 2008. Understanding the value of features for coreference resolution. In Proceedings of the 2008 Conference on Empirical Methods in Natural Language Processing, pages 294– 303. [find](https://scholar.google.com/scholar?q=Eric+Bengtson+and+Dan+Roth%2E+2008%2E+Understanding+the+value+of+features+for+coreference+resolution%2E+In+Proceedings+of+the+2008+Conference+on+Empirical+Methods+in+Natural+Language+Processing%2C+pages+294%E2%80%93+) [our card](https://md.research.elamur.ai/free/papers/bengtson-2008-understanding-value-features-coreference-resolution/) 3. Shuang Chen, Jinpeng Wang, Feng Jiang, and Chin-Yew Lin. 2020. Improving entity linking by modeling latent entity type information. arXiv preprint arXiv:2001.01447. — [arXiv:2001.01447](https://arxiv.org/abs/2001.01447) [find](https://scholar.google.com/scholar?q=Shuang+Chen%2C+Jinpeng+Wang%2C+Feng+Jiang%2C+and+Chin%2DYew+Lin%2E+2020%2E+Improving+entity+linking+by+modeling+latent+entity+type+information%2E+arXiv+preprint+arXiv%3A2001%2E01447%2E+%E2%80%94+%5BarXiv%3A2001%2E01447%5D%28https%3A%2F%2Farxiv%2E) 4. H. H. Clark. 1975. Bridging. In Proceedings of TIN-LAP. \#\# Conclusion Kevin Clark and Christopher D Manning. 2016. Deep reinforcement learning for mention-ranking coreference models. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 2256–2262. [find](https://scholar.google.com/scholar?q=H%2E+H%2E+Clark%2E+1975%2E+Bridging%2E+In+Proceedings+of+TIN%2DLAP%2E+%5C%23%5C%23+Conclusion+Kevin+Clark+and+Christopher+D+Manning%2E+2016%2E+Deep+reinforcement+learning+for+mention%2Dranking+coreference+models%2E+In+Proceedings+)