Elamur Research⌕
☰
home / papers / 2021 / Stay Together: A System for Single and Split-antecedent Anaphora Resolution

Stay Together: A System for Single and Split-antecedent Anaphora Resolution

Summary

Split-antecedent anaphora occurs when a plural expression refers to a set built from separately mentioned entities, as in "they" referring to John and Mary. It is rare, often left unannotated, and earlier systems handled it in isolation, assuming gold mentions and/or gold anaphors. This paper presents a single system that resolves both ordinary single-antecedent anaphors and split-antecedent ones from predicted mentions. The authors extend the cluster-ranking coreference model of Yu et al. (2020b), the state of the art on ARRAU, which already handles singletons and non-referring expressions. The new module reuses that model's mention representations and candidate clusters. It treats each mention–cluster pair as a binary decision about whether the cluster is one of the mention's antecedents. The module is trained either JOINT with the base model or on top of a PRE-TRAINED one. They also propose an extended LEA metric that explicitly represents plural entities, so both anaphora types are scored together. Both variants clearly beat heuristic and random baselines. In the gold setting, the PRE-TRAINED variant also outperforms the earlier dedicated split-antecedent system.

Key points

  • Data and code: Experiments use the ARRAU corpus with the standard RST split. Code is released at github.com/juntaoy/dali-full-anaphora.
  • JOINT vs PRE-TRAINED: The two setups perform equally on split antecedents, but PRE-TRAINED is better on the other aspects of anaphoric interpretation.
  • Predicted-mention bottleneck:
    • 98.33% of split-antecedent anaphors appear among predicted mentions, but only 65% among the discourse-new mentions the system considers.
    • Overall, about 45% of anaphors are recognized.
    • Lenient F1 is 36.4% with predicted anaphors and 47.7% when the recalled anaphors are supplied, compared with 68.1% in the fully gold setting.
    • Noisy mentions and clusters explain about two thirds of that gap.
  • Gold-setting comparison with Yu et al. (2020a):
    • Anaphors with 3+ antecedents: lenient 61.0 vs 52.5, strict 14.3 vs 0.0.
    • Non-pronominal anaphors: lenient 62.9 vs 57.5, strict 38.9 vs 19.4.
    • Pronouns: the older system remains better (lenient 82.4 vs 76.4).
  • Evaluation: The extended LEA metric avoids the flawed practice of copying a plural mention into each antecedent's cluster, which is how the CoNLL scorer was used in earlier work.

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

Topics: Coreference resolution

Cited by

9 papers of our library cite it (of the 140673 whose bibliographies we read); 6 of them have a page here, 3 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2021.naacl-main.329 · arXiv 2104.05320

@inproceedings{yu2021stay,
  title = {{Stay Together: A System for Single and Split-antecedent Anaphora Resolution}},
  author = {Yu, Juntao and Moosavi, Nafise Sadat and Paun, Silviu and Poesio, Massimo},
  booktitle = {Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies},
  year = {2021},
  doi = {10.18653/v1/2021.naacl-main.329},
  eprint = {2104.05320},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/2021.naacl-main.329}
}

References

33 references, 25 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Catalina Barbu and Ruslan Mitkov. 2001. Evaluation tool for rule-based anaphora resolution methods. In Proceedings of the 39th Annual Meeting of the Association for Computational Linguistics, pages 34–41, Toulouse, France. Association for Computational Linguistics. find our card
  2. Kevin Clark and Christopher D. Manning. 2016a. Deep reinforcement learning for mention-ranking coreference models. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 2256–2262, Austin, Texas. Association for Computational Linguistics. find our card
  3. In this paper, we introduced a neural system per- Kevin Clark and Christopher D. Manning. 2016b. Im-forming both single and split-antecedent anaphora proving coreference resolution by learning entity-4182 level distributed representations. In Proceedings of the 54th Annual Meeting of the Association… find
  4. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… find
  5. Yufang Hou. 2020. Bridging anaphora resolution as question answering. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1428–1438, Online. Association for Computational Linguistics. pages 188–197, Copenhagen, Denmark. Association for Computational Ling… find our card
  6. Kenton Lee, Luheng He, and Luke Zettlemoyer. 2018. Higher-order coreference resolution with coarse-to-fine inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers), pages 687–… find our card
  7. Quan Liu, Hui Jiang, Andrew Evdokimov, Zhen-Hua Ling, Xiaodan Zhu, Si Wei, and Yu Hu. 2017. Cause-effect knowledge acquisition and neural association model for solving a set of winograd schema problems. In Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence, IJC… find
  8. Mandar Joshi, Danqi Chen, Yinhan Liu, Daniel S. Weld, Luke Zettlemoyer, and Omer Levy. 2020. SpanBERT: Improving pre-training by representing and predicting spans. Transactions of the Association for Computational Linguistics, 8:64–77. find our card
  9. Ana Marasović, Leo Born, Juri Opitz, and Anette Frank. 2017. A mention-ranking model for abstract anaphora resolution. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 221–232, Copenhagen, Denmark. Association for Computational Linguistics. find our card
  10. 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… find our card
  11. Nafise Sadat Moosavi and Michael Strube. 2016. Which coreference evaluation metric do you trust? a proposal for a link-based entity aware metric. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 632–642, Berlin, Germany. Associ… find our card
  12. Ben Kantor and Amir Globerson. 2019. Coreference resolution with entity equalization. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 673–677, Florence, Italy. Association for Computational Linguistics. find our card
  13. Varada Kolhatkar and Graeme Hirst. 2014. Resolving shell nouns. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 499–510, Doha, Qatar. Association for Computational Linguistics. find
  14. Varada Kolhatkar, Adam Roussel, Stefanie Dipper, and Heike Zinsmeister. 2018. Anaphora with non-nominal antecedents in computational linguistics: a Survey. Computational Linguistics, 44(3):547–612. find our card
  15. Heeyoung. Lee, Angel. Chang, Yves. Peirsman, Nathaneal. Chambers, Mihai. Surdeanu, and Dan. Jurafsky. 2013. Deterministic coreference resolution based on entity-centric, precision-ranked rules. Computational Linguistics, 39(4):885–916. find our card
  16. 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, Haoruo Peng, Daniel Khashabi, and Dan Roth. 2015. Solving hard coreference problems. In Proceedings of t… find our card
  17. Jeffrey Pennington, Richard Socher, and Christopher Manning. 2014. GloVe: Global vectors for word representation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1532–1543, Doha, Qatar. Association for Computational Linguistics. find
  18. Matthew Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word representations. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technolog… find
  19. Massimo Poesio and Ron Artstein. 2008. Anaphoric annotation in the ARRAU corpus. In Proceedings of the Sixth International Conference on Language Resources and Evaluation (LREC’08), Marrakech, Morocco. European Language Resources Association (ELRA). find our card
  20. Massimo Poesio, Yulia Grishina, Varada Kolhatkar, Nafise Moosavi, Ina Roesiger, Adam Roussel, Fabian Simonjetz, Alexandra Uma, Olga Uryupina, Juntao Yu, and Heike Zinsmeister. 2018. Anaphora resolution with the ARRAU corpus. In Proceedings of the First Workshop on Computational Models of Reference,… find our card
  21. Sameer Pradhan, Alessandro Moschitti, Nianwen Xue, Olga Uryupina, and Yuchen Zhang. 2012. CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes. In Proceedings of the Sixteenth Conference on Computational Natural Language Learning (CoNLL 2012), Jeju, Korea. find our card
  22. Altaf Rahman and Vincent Ng. 2012. Resolving complex cases of definite pronouns: The Winograd schema challenge. In Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, pages 777–789, Jeju Island, Korea. Association… find our card
  23. Keisuke Sakaguchi, Ronan Le Bras, Chandra Bhagavatula, and Yejin Choi. 2020. Winogrande: An adversarial winograd schema challenge at scale. Proceedings of the AAAI Conference on Artificial Intelligence, 34:8732–8740. find
  24. Olga Uryupina, Ron Artstein, Antonella Bristot, Federica Cavicchio, Francesca Delogu, Kepa J. Rodriguez, and Massimo Poesio. 2020. Annotating a broad range of anaphoric phenomena, in a variety of genres: the ARRAU corpus. Journal of Natural Language Engineering. find
  25. Hardik Vala, Andrew Piper, and Derek Ruths. 2016. The more antecedents, the merrier: Resolving multi-antecedent anaphors. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2287– 2296, Berlin, Germany. Association for Computation… find our card
  26. Sam Wiseman, Alexander M. Rush, and Stuart M. Shieber. 2016. Learning global features for coreference resolution. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 994–1004, San Diego, California.… find our card
  27. Wei Wu, Fei Wang, Arianna Yuan, Fei Wu, and Jiwei Li. 2020. CorefQA: Coreference resolution as query-based span prediction. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6953–6963, Online. Association for Computational Linguistics. find our card
  28. Juntao Yu, Nafise Sadat Moosavi, Silviu Paun, and Massimo Poesio. 2020a. Free the plural: Unrestricted split-antecedent anaphora resolution. In Proceedings of the 28th International Conference on Computational Linguistics, pages 6113–6125, Barcelona, Spain (Online). International Committee on Comput… find our card
  29. Juntao Yu and Massimo Poesio. 2020. Multitask learning based neural bridging reference resolution. In Proceedings of the 28th International Conference on Computational Linguistics, pages 3534–3546, Barcelona, Spain (Online). International Committee on Computational Linguistics. find our card
  30. Juntao Yu, Alexandra Uma, and Massimo Poesio. 2020b. A cluster ranking model for full anaphora resolution. In Proceedings of the 12th Language Resources and Evaluation Conference, pages 11–20, Marseille, France. European Language Resources Association. find our card
  31. Ethan Zhou and Jinho D. Choi. 2018. They exist! introducing plural mentions to coreference resolution and entity linking. In Proceedings of the 27th International Conference on Computational Linguistics, pages 24–34, Santa Fe, New Mexico, USA. Association for Computational Linguistics. find our card
  32. Kellie Webster, Marta Recasens, Vera Axelrod, and Jason Baldridge. 2018. Mind the GAP: A balanced corpus of gendered ambiguous pronouns. Transactions of the Association for Computational Linguistics, 6:605–617. find our card
  33. Sam Wiseman, Alexander M. Rush, Stuart Shieber, and Jason Weston. 2015. Learning anaphoricity and antecedent ranking features for coreference resolution. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural L… find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/yu-2021-stay-together-system-single-split/