Supervised Models for Coreference Resolution

Altaf Rahman and Vincent Ng.
Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pp. 968-977, 2009.

Click here for the PostScript or PDF version. The talk slides are available here.


Traditional learning-based coreference resolvers operate by training a mention-pair classifier for determining whether two mentions are coreferent or not. Two independent lines of recent research have attempted to improve these mention-pair classifiers, one by learning a mention-ranking model to rank preceding mentions for a given anaphor, and the other by training an entity-mention classifier to determine whether a preceding cluster is coreferent with a given mention. We propose a cluster-ranking approach to coreference resolution that combines the strengths of mention rankers and entity-mention models. We additionally show how our cluster-ranking framework naturally allows anaphoricity determination to be learned jointly with coreference resolution. Experimental results on the ACE data sets demonstrate its superior performance to competing approaches.


CherryPicker, a coreference tool that implements the cluster-ranking model described in this paper, is available from this page.

Train-test split

Here are the lists of names of the files from the English portion of the ACE 2005 Multilingual Training Corpus we used for training and testing.

BibTeX entry

  author = {Altaf Rahman and Vincent Ng},
  title = {Supervised Models for Coreference Resolution},
  booktitle = {Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing},
  pages = {968--977},
  year = 2009