Natural Language Understanding Wiki

Lee et al. (2012)[1]

From Lee et al. (2012)[1]: "limited work that incorporates event-related information in entity coreference, typically by incorporating the verbs in context as features. For instance, Haghighi and Klein (2010) include the governor of the head of nominal mentions as features in their model. Rahman and Ng (2011) also used event-related information by looking at which semantic role the entity mentions can have and the verb pairs of their predicates.

Humphreys et al. (1997) considered entities as well as events, but due to the lack of a corpus annotated with event coreference, their approach was only evaluated implicitly in the MUC-6 template filling task. To our knowledge, the only previous work that considered entity and event coreference resolution jointly is He (2007), but limited to the medical domain and focused on just five semantic categories."


Lee et al. (2012)[1] extended the ECB corpus with entity coreference information. The resulting corpus can be downloaded at: It is sometimes referred to as "ECB 0.1" (Cybulska and Vossen, 2014[2], p. 4549).

Statistics of Lee et al.'s corpus (Table 1 in Lee et al. (2012))
Training Dev Test Total
# Topics 12 3 28 43
# Documents 112 39 331 482
# Entities 459 46 563 1068
# Entity Mentions 1723 259 3465 5447
# Events 300 30 444 774
# Event Mentions 751 140 1642 2533


  1. 1.0 1.1 1.2 Lee, H., Recasens, M., Chang, A., Surdeanu, M., & Jurafsky, D. (2012). Joint Entity and Event Coreference Resolution across Documents. (EMNLP-CoNLL 2012) Proceedings of the 2012 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning, (July), 489–500. Retrieved from
  2. Cybulska, A., & Vossen, P. (2014). Using a sledgehammer to crack a nut? Lexical diversity and event coreference resolution. Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC’14), 4545–4552. Retrieved from