Elamur Research⌕
☰
home / papers / 2015 / A Joint Framework for Coreference Resolution and Mention Head Detection

A Joint Framework for Coreference Resolution and Mention Head Detection

Summary

Coreference systems lose much of their accuracy when they must find mentions in raw text rather than receive gold-annotated ones. This end-to-end setting is the one that matters in practice. Peng, Chang and Roth address this gap by not treating mention detection as a fixed preprocessing stage. Their system works with mention heads, which are shorter, never overlap, and carry most of the information that coreference features use. They first generate high-recall head candidates by combining a statistical component with existing lexical resources. An Integer Linear Program then jointly decides which candidates are real heads and how they link into coreference chains. It learns a head classifier and a left-linking mention-pair scorer at the same time. Full mention spans are recovered only after linking, using a binary classifier. Experiments on ACE-2004 and CoNLL-2012 (OntoNotes-5.0) show large gains in mention detection and the best end-to-end coreference results reported at the time. The work argues for evaluating NLP systems on raw input.

Key points

  • Existing systems, including the Illinois, Berkeley and Stanford coreference systems, score more than 10 points lower with predicted mentions than with gold ones. For example, Illinois on CoNLL-12 falls from 77.05 to 60.00.
  • The joint ILP extends the best-left-link formulation of Chang et al. (2011) with a variable for each head candidate. A link is allowed only if the candidate is accepted, so the model can discard spurious mentions and singletons. This matters for OntoNotes, which drops singleton mentions.
  • Mention detection improves by roughly 10–15 F1. The end-to-end system (H-Joint-M) reduces mention detection error by 55.36% on ACE-2004 and 35.49% on CoNLL-2012, measured on full mentions.
  • The gap between coreference on gold and predicted mentions shrinks by over 20%: 26.28% on ACE-2004 and 22.60% on CoNLL-2012. The candidate generator and the joint inference contribute about equally to this reduction.
  • Because OntoNotes has no head annotations, heads were extracted in ACE style. The software is released through the Illinois CogComp group.

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

27 papers of our library cite it (of the 140673 whose bibliographies we read); 11 of them have a page here, 16 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/k15-1002

@inproceedings{peng2015joint,
  title = {{A Joint Framework for Coreference Resolution and Mention Head Detection}},
  author = {Peng, Haoruo and Chang, Kai-Wei and Roth, Dan},
  booktitle = {Proceedings of the Nineteenth Conference on Computational Natural Language Learning},
  year = {2015},
  doi = {10.18653/v1/k15-1002},
  url = {https://doi.org/10.18653/v1/k15-1002}
}

References

35 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. A. Bagga and B. Baldwin. 1998. Algorithms for scoring coreference chains. In In The First International Conference on Language Resources and Evaluation Workshop on Linguistics Coreference. find our card
  2. E. Bengtson and D. Roth. 2008. Understanding the value of features for coreference resolution. In EMNLP. find our card
  3. A. Björkelund and J Kuhn. 2014. Learning structured Perceptrons for coreference resolution with latent antecedents and non-local features. In ACL. find our card
  4. C. Cardie and D. Pierce. 1998. Error-driven pruning of Treebanks grammars for base noun phrase identification. In Proceedings of ACL-98. find
  5. K. Chang, R. Samdani, A. Rozovskaya, N. Rizzolo, M. Sammons, and D. Roth. 2011. Inference protocols for coreference resolution. In CoNLL. find our card
  6. K.-W. Chang, R. Samdani, and D. Roth. 2013. A constrained latent variable model for coreference resolution. In EMNLP. find our card
  7. K. Church. 1988. A stochastic parts program and noun phrase parser for unrestricted text. In ANLP. find
  8. M. Collins. 1999. Head-driven Statistical Models for Natural Language Parsing. Ph.D. thesis, Computer Science Department, University of Pennsylvenia. find
  9. D. Connolly, J. Burger, and D. Day. 1997. A machine learning approach to anaphoric reference. In New Methods in Language Processing. find
  10. A. Culotta, M. Wick, and A. McCallum. 2007. First-order probabilistic models for coreference resolution. In NAACL. find our card
  11. P. Denis and J. Baldridge. 2007. Joint determination of anaphoricity and coreference resolution using integer programming. In NAACL. find our card
  12. G. Durrett and D. Klein. 2013. Easy victories and uphill battles in coreference resolution. In EMNLP. find our card
  13. G. Durrett and D. Klein. 2014. A joint model for entity analysis: Coreference, typing, and linking. find our card
  14. J. Finkel and C. Manning. 2009. Nested named entity recognition. In EMNLP. find
  15. J. R. Hobbs. 1978. Resolving pronoun references. Lingua. find our card
  16. E. Hovy, M. Marcus, M. Palmer, L. Ramshaw, and R. Weischedel. 2006. Ontonotes: The 90% solution. In Proceedings of HLT/NAACL. find
  17. J. Jackson and M. Craven. 1996. Learning sparse perceptrons. Proceedings of the 1996 Advances in Neural Information Processing Systems. find
  18. S. Lappin and H. Leass. 1994. An algorithm for pronominal anaphora resolution. Computational linguistics. find our card
  19. E. Lassalle and P. Denis. 2015. Joint anaphoricity detection and coreference resolution with constrained latent structures. find
  20. H. Lee, Y. Peirsman, A. Chang, N. Chambers, M. Surdeanu, and D. Jurafsky. 2011. Stanford’s multi-pass sieve coreference resolution system at the conll-2011 shared task. In Proceedings of the CoNLL-2011 Shared Task. find our card
  21. H. Lee, M. Recasens, A. Chang, M. Surdeanu, and D. Jurafsky. 2012. Joint entity and event coreference resolution across documents. In EMNLP. find our card
  22. X. Luo. 2005. On coreference resolution performance metrics. In EMNLP. find our card
  23. V. Ng and C. Cardie. 2002. Improving machine learning approaches to coreference resolution. In ACL. find our card
  24. NIST. 2004. The ACE evaluation plan. find
  25. S. Pradhan, A. Moschitti, N. Xue, O. Uryupina, and Y. Zhang. 2012. CoNLL-2012 shared task: Modeling multilingual unrestricted coreference in OntoNotes. In CoNLL. find our card
  26. V. Punyakanok and D. Roth. 2001. The use of classifiers in sequential inference. In NIPS. find
  27. L. R. Rabiner. 1989. A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE. find
  28. L. A. Ramshaw and M. P. Marcus. 1995. Text chunking using transformation-based learning. In Proceedings of the Third Annual Workshop on Very Large Corpora. find
  29. L. Ratinov and D. Roth. 2009. Design challenges and misconceptions in named entity recognition. In Proc. of the Annual Conference on Computational Natural Language Learning (CoNLL). find our card
  30. M. Recasens, M.-C. de Marneffe, and C. Potts. 2013. The life and death of discourse entities: Identifying singleton mentions. In NAACL. find
  31. D. Roth and W. Yih. 2004. A linear programming formulation for global inference in natural language tasks. In Hwee Tou Ng and Ellen Riloff, editors, CoNLL. find our card
  32. S. Sarawagi and W. Cohen. 2004. Semi-Markov conditional random fields for information extraction. In NIPS. find
  33. Y. Song, J. Jiang, W.-X. Zhao, S. Li, and H. Wang. 2012. Joint learning for coreference resolution with markov logic. In Proceedings of the 2012 Joint Conference of EMNLP-CoNLL. find
  34. W. M. Soon, H. T. Ng, and D. C. Y. Lim. 2001. A machine learning approach to coreference resolution of noun phrases. Comput. Linguist. find our card
  35. M. Vilain, J. Burger, J. Aberdeen, D. Connolly, and L. Hirschman. 1995. A model-theoretic coreference scoring scheme. In Proceedings of the 6th conference on Message understanding. find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/papers/peng-2015-joint-framework-coreference-resolution-mention/