Elamur Research⌕
☰
home / papers / 2010 / A Global Relaxation Labeling Approach to Coreference Resolution

A Global Relaxation Labeling Approach to Coreference Resolution

Summary

The paper addresses coreference resolution, which is grouping the mentions in a text that refer to the same entity. Typical systems first score mention pairs or groups and then build chains in a separate step. Errors from the context-poor first step carry over into the second. The authors merge both steps into one global procedure. Each mention becomes a vertex in a graph, and the edge weights are sums of weighted constraints. Constraints are learned automatically from a training corpus using pairwise feature functions, with separate models depending on anaphor type. The partitioning of vertices into entities is then found iteratively by relaxation labeling. Each vertex holds a probability distribution over possible partitions, which is updated from the support of earlier mentions. This keeps the result consistent without explicit transitivity constraints. Pruning, reordering, and an informed initial state further improve results. On the ACE-phase02 corpus, the method beats decision-tree baselines, an ILP chain-formation baseline, and the best published systems using the same data and metrics. It also scales better than ILP and makes it easy to add knowledge from other sources as constraints.

Key points

  • Best configuration (Relax with pruning and reordering) reaches a global CEAF F1 of 69.7 and a B3 F1 of 74.9 on ACE-phase02 (bnews, npaper, nwire).
  • Prior results on the same data: MaxEnt+ILP (Denis, 2007) scored 66.2 CEAF / 72.7 B3, and Rankers scored 67.0 CEAF / 72.7 B3.
  • Hill Climbing feature selection, optimized for CEAF, raised the decision-tree baseline by 5.1 CEAF and 5.3 B3 points.
  • The ILP baseline needs a cubic number of triangular constraints, so it could only handle documents with fewer than 200 mentions.
  • Relaxation stops when label probabilities change by less than 0.001 or after 2000 iterations. An informed initial state outperformed a random one (averaged over 3 runs).

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

6 papers of our library cite it (of the 140673 whose bibliographies we read); each has a page here.

Cite

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

@inproceedings{sapena2010global,
  title = {{A Global Relaxation Labeling Approach to Coreference Resolution}},
  author = {Sapena, Emili and Padró, Lluís and Turmo, Jordi},
  booktitle = {Coling 2010: Posters},
  year = {2010},
  url = {https://aclanthology.org/C10-2125/}
}

References

31 references, 16 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Aone, C. and S.W. Bennett. 1995. Evaluating automated and manual acquisition of anaphora resolution strategies. In Proceedings of the 33rd annual meeting on ACL, pages 122–129. find our card
  2. Atserias, J., B. Casas, E. Comelles, M. González, L. Padró, and M. Padró. 2006. Freeling 1.3: Syntactic and semantic services in an open-source nlp library. In Proceedings of the fifth international conference on Language Resources and Evaluation (LREC 2006), ELRA. Genoa, Italy. find
  3. Atserias, J. 2006. Towards Robustness in Natural Language Understanding. Ph.D. Thesis, Dept. Lenguajes y Sistemas Informáticos. Euskal Herriko Unibertsitatea. Donosti. Spain. find
  4. Bagga, A. and B. Baldwin. 1998. Algorithms for scoring coreference chains. Proceedings of the Linguistic Coreference Workshop at LREC, pages 563–566. find our card
  5. Culotta, A., M. Wick, and A. McCallum. 2007. First-Order Probabilistic Models for Coreference Resolution. Proceedings of NAACL HLT, pages 81–88. find our card
  6. Denis, P. and J. Baldridge. 2007. Joint Determination of Anaphoricity and Coreference Resolution using Integer Programming. Proceedings of NAACL HLT, pages 236– 243. find our card
  7. Denis, P. and J. Baldridge. 2008. Specialized models and ranking for coreference resolution. Proceedings of the EMNLP, Hawaii, USA. find our card
  8. Denis, P. 2007. New Learning Models for Robust Reference Resolution. Ph.D. dissertation, University of Texas at Austin. find
  9. Finkel, J.R. and C.D. Manning. 2008. Enforcing transitivity in coreference resolution. In Proceedings of the 46th Annual Meeting of the ACL HLT: Short Papers, pages 45–48. Association for Computational Linguistics. find our card
  10. Gimenez, J. and L. Marquez. 2004. Svmtool: A general pos tagger generator based on support vector machines. In Proceedings of the 4th International Conference on Language Resources and Evaluation, pages 43–46. find our card
  11. Hoste, V. 2005. Optimization Issues in Machine Learning of Coreference Resolution. PhD thesis. find
  12. Hummel, R. A. and S. W. Zucker. 1987. On the foundations of relaxation labeling processes. pages 585–605. find
  13. Klenner, M. and É. Ailloud. 2008. Enhancing Coreference Clustering. In Proceedings of the Second Workshop on Anaphora Resolution. WAR II. find
  14. Klenner, M. and E. Ailloud. 2009. Optimization in Coreference Resolution Is Not Needed: A Nearly-Optimal Algorithm with Intensional Constraints. In Proceedings of the 12th Conference of the EACL. find our card
  15. Klenner, M. 2007. Enforcing consistency on coreference sets. In Recent Advances in Natural Language Processing (RANLP), pages 323–328. find
  16. Luo, X., A. Ittycheriah, H. Jing, N. Kambhatla, and S. Roukos. 2004. A mention-synchronous coreference resolution algorithm based on the bell tree. In Proceedings of 42nd ACL, page 135. find our card
  17. Luo, X. 2005. On coreference resolution performance metrics. Proc. of HLT-EMNLP, pages 25–32. find our card
  18. Luo, X. 2007. Coreference or not: A twin model for coreference resolution. In Proceedings of NAACL HLT, pages 73–80. find our card
  19. Màrquez, L., L. Padró, and H. Rodrı́guez. 2000. A machine learning approach for pos tagging. Machine Learning Journal, 39(1):59–91. find
  20. McCallum, A. and B. Wellner. 2005. Conditional models of identity uncertainty with application to noun coreference. Advances in Neural Information Processing Systems, 17:905–912. find
  21. McCarthy, J.F. and W.G. Lehnert. 1995. Using decision trees for coreference resolution. Proceedings of the Fourteenth International Conference on Artificial Intelligence, pages 1050–1055. find
  22. Ng, V. and C. Cardie. 2002. Improving machine learning approaches to coreference resolution. Proceedings of the 40th Annual Meeting on Association for Computational Linguistics, pages 104–111. find our card
  23. Nicolae, C. and G. Nicolae. 2006. Best Cut: A Graph Algorithm for Coreference Resolution. Proceedings of the 2006 Conference on EMNLP, pages 275–283. find
  24. Pelillo, M. 1997. The dynamics of nonlinear relaxation labeling processes. Journal of Mathematical Imaging and Vision, 7(4):309–323. find
  25. Quinlan, J.R. 1993. C4.5: Programs for Machine Learning. Morgan Kaufmann. find our card
  26. Rosenfeld, R., R. A. Hummel, and S. W. Zucker. 1976. Scene labelling by relaxation operations. IEEE Transactions on Systems, Man and Cybernetics, 6(6):420–433. find
  27. Soon, W.M., H.T. Ng, and D.C.Y. Lim. 2001. A Machine Learning Approach to Coreference Resolution of Noun Phrases. Computational Linguistics, 27(4):521–544. find our card
  28. Surdeanu, M., J. Turmo, and E. Comelles. 2005. Named Entity Recognition from Spontaneous Open-Domain Speech. In Ninth European Conference on Speech Communication and Technology. ISCA. find
  29. Torras, C. 1989. Relaxation and neural learning: Points of convergence and divergence. Journal of Parallel and Distributed Computing, 6:217–244. find
  30. Vilain, M., J. Burger, J. Aberdeen, D. Connolly, and L. Hirschman. 1995. A model-theoretic coreference scoring scheme. Proceedings of the 6th conference on Message understanding, pages 45–52. find our card
  31. Yang, X., G. Zhou, J. Su, and C.L. Tan. 2003. Coreference resolution using competition learning approach. In ACL ’03: Proceedings of the 41st Annual Meeting on Association for Computational Linguistics, pages 176–183. find our card

↑ back to top