Elamur Research⌕
☰
home / papers / 2011 / An Incremental Entity-Mention Model for Coreference Resolution with Restrictive Antecedent Accessibility

An Incremental Entity-Mention Model for Coreference Resolution with Restrictive Antecedent Accessibility

Summary

The paper addresses weaknesses of the classic mention-pair approach to coreference resolution. That approach classifies every anaphor–antecedent pair before building clusters, which produces redundant pairs, heavily skewed training data, underspecified candidates, and no built-in transitivity.

Klenner and Tuggener propose an incremental entity-mention model. Markables are processed in textual order, and each new mention is compared only against a "virtual prototype" of every emerging coreference set, which pools the morphological and semantic properties of its members, plus a buffer of non-anaphoric mentions. Access to the buffer is restricted, and further filters handle anaphora type, an approximation of binding theory via subclause detection, and WordNet/GermaNet lookups. Antecedents are chosen either by TiMBL classifiers or by a Lappin–Leass-style salience measure whose dependency-label weights are estimated from gold-standard data (TüBa-D/Z, OntoNotes).

On German, the incremental model outperforms a non-incremental counterpart. The simple salience variant beats the SemEval 2010 systems in a post-task evaluation and is competitive in BioNLP 2011. This suggests that improving pair generation and antecedent accessibility matters as much as improving the classifier.

Key points

  • Training instances on fold 1 of TüBa-D/Z (about 5,000 sentences) drop roughly fourfold, from 171,526 (mention-pair) to 40,229 (entity-mention).
  • Separate classifiers are trained per anaphora type (nominal, reflexive, possessive, relative, personal pronouns) using Soon et al. (2001)-style features. The mention-pair baseline clusters with the Balas algorithm.
  • In the SemEval 2010 open/regular setting, the salience-based incremental system scored best on nearly all metrics, the exception being MUC for English. For German, its CEAF reached R 76.8 / P 70.4, against BART's 61.4 / 61.2.
  • The gains came mainly from precision. Recall stayed relatively low partly because bridging anaphora was not resolved.
  • On the CoNLL 2011 development set the system scored MUC F1 47.39, BCUB 65.44, CEAFE 42.03, and BLANC 66.81. The authors attribute the lower scores mainly to the stricter CoNLL scorer and to singletons not being annotated.

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

8 papers of our library cite it (of the 140673 whose bibliographies we read); 4 of them have a page here, 4 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{klenner2011incremental,
  title = {{An Incremental Entity-Mention Model for Coreference Resolution with Restrictive Antecedent Accessibility}},
  author = {Klenner, Manfred and Tuggener, Don},
  booktitle = {Proceedings of the International Conference Recent Advances in Natural Language Processing 2011},
  year = {2011},
  doi = {10.5167/uzh-49838},
  url = {https://doi.org/10.5167/uzh-49838}
}

References

24 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Daniel Büring. 2005. Binding Theory. Cambridge Textbooks in Linguistics. Cambridge University Press, Cambridge. find
  2. Jie Cai and Michael Strube. 2010. Evaluation metrics for end-to-end coreference resolution systems. In Proceedings of the 11th Annual Meeting of the Special Interest Group on Discourse and Dialogue, SIGDIAL ’10, pages 28–36, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  3. Aron Culotta, Michael Wick, and Andrew McCallum. 2007. First-order probabilistic models for coreference resolution. In HLT ’07: The Conference of the North American Chapter of ACL; Proceedings of the Main Conference, pages 81–88, Rochester, New York, April. Association for Computational Linguistics.… find our card
  4. Walter Daelemans, Jakub Zavrel, Ko van der Sloot, and Antal van den Bosch. 2007. Timbl: Tilburg memory-based learner. Technical report, Induction of Linguistic Knowledge, Tilburg University and CNTS Research Group, University of Antwerp. find
  5. Hal Daume III and Daniel Marcu. 2005. A large-scale exploration of effective global features for a joint entity detection and tracking model. In HLT ’05: Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 97–104, Morristown, NJ, USA… find
  6. Pascal Denis and Jason Baldridge. 2009. Global joint models for coreference resolution and named entity classification. In Procesamiento del Lenguaje Natural 42, pages 87–96, Barcelona: SEPLN. find
  7. Christiane Fellbaum. 1998. WordNet: An Electronic Lexical Database (Language, Speech, and Communication). The MIT Press, May. find
  8. Jenny Rose Finkel and Christopher D. Manning. 2008. Enforcing transitivity in coreference resolution. In HLT ’08: Proceedings of the 46th Annual Meeting of the ACL on Human Language Technologies, pages 45–48, Morristown, NJ, USA. Association for Computational Linguistics. find our card
  9. Niye Ge, John Hale, and Eugene Charniak. 1998. A statistical approach to anaphora resolution. In Proceedings of the Sixth Workshop on Very Large Corpora, pages 161–171, Montreal, Canada. find our card
  10. Birgit Hamp and Helmut Feldweg. 1997. GermaNet – a lexical-semantic net for German. In Proceedings of ACL workshop Automatic Information Extraction and Building of Lexical Semantic Resources for NLP Applications, pages 9–15, Somerset, NJ, USA. Association for Computational Linguistics. find our card
  11. Manfred Klenner. 2007. Enforcing consistency on coreference sets. In Proceedings of the International Conference on Recent Advances in Natural Language Processing, pages 323–328, Borovets, Bulgaria. find
  12. Shalom Lappin and Herbert J Leass. 1994. An algorithm for pronominal anaphora resolution. Computational Linguistics, 20:535–561. find our card
  13. Xiaoqiang Luo, Abe Ittycheriah, Hongyan Jing, Nanda Kambhatla, and Salim Roukos. 2004. A mention-synchronous coreference resolution algorithm based on the bell tree. In Proceedings of the 42nd Annual Meeting of ACL, ACL ’04, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  14. Xiaoqiang Luo. 2005. On coreference resolution performance metrics. In HLT ’05: Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 25–32, Morristown, NJ, USA. Association for Computational Linguistics. find our card
  15. Karin Naumann, 2006. Manual for the Annotation of Indocument Referential Relations. SFS (Seminar für Sprachwissenschaft), http://www.sfs.uni-tuebingen.de/tuebadz.shtml. ↗
  16. Vincent Ng. 2010. Supervised noun phrase coreference research: the first fifteen years. In Proceedings of the 48th Annual Meeting of ACL, ACL ’10, pages 1396–1411, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  17. Karthik Raghunathan, Heeyoung Lee, Sudarshan Rangarajan, Nathanael Chambers, Mihai Surdeanu, Dan Jurafsky, and Christopher Manning. 2010. A multi-pass sieve for coreference resolution. In Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, EMNLP ’10, pages 492–501… find our card
  18. Altaf Rahman and Vincent Ng. 2009. Supervised models for coreference resolution. In Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing: Volume 2 - Volume 2, EMNLP ’09, pages 968–977, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card
  19. Marta Recasens, Lluı́s Màrquez, Emili Sapena, M. Antònia Martı́, Mariona Taulé, Véronique Hoste, Massimo Poesio, and Yannick Versley. 2010. Semeval-2010 task 1: Coreference resolution in multiple languages. In Proceedings of the 5th International Workshop on Semantic Evaluation, SemEval ’10, pag… find our card
  20. Gerold Schneider. 2008. Hybrid Long-Distance Functional Dependency Parsing. Doctoral Thesis, Institute of Computational Linguistics, Univ. of Zurich. find
  21. Rico Sennrich, Gerold Schneider, Martin Volk, and Martin Warin. 2009. A New Hybrid Dependency Parser for German. In Proc. of the German Society for Computational Linguistics and Language Technology 2009 (GSCL 2009), pages 115–124, Potsdam, Germany. find
  22. Wee M. Soon, Hwee T. Ng, and Daniel. 2001. A machine learning approach to coreference resolution of noun phrases. Computational Linguistics, 27(4):521–544, December. find our card
  23. Holger Wunsch, Sandra Kübler, and Rachael Cantrell. 2009. Instance sampling methods for pronoun resolution. In Proceedings of Proceedings of the International Conference on Recent Advances in Natural Language Processing, Borovets, Bulgaria. find
  24. Xiaofeng Yang, Jian Su, Guodong Zhou, and Chew Lim Tan. 2004. An np-cluster based approach to coreference resolution. In Proceedings of the 20th International Conference on Computational Linguistics, COLING ’04, Stroudsburg, PA, USA. Association for Computational Linguistics. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/klenner-2011-incremental-entity-mention-model-coreference/