Triad-based Neural Network for Coreference Resolution

We propose a triad-based neural network system that generates affinity scores between entity mentions for coreference resolution. The system simultaneously accepts three mentions as input, taking mutual dependency and logical constraints of all three mentions into account, and thus makes more accurate predictions than the traditional pairwise approach. Depending on system choices, the affinity scores can be further used in clustering or mention ranking.

Fig.1. Clustering results from the test file. Top subfigure is the true clusters. Bottom is the predicted results. Mentions with the same color are in the same cluster.


Y. Meng A. Rumshisky Triad-based Neural Network for Coreference Resolution. Proceedings of COLING. 2018

@article{meng_triad-based_2018, author = "Meng, Yuanliang and Rumshisky, Anna", title = "Triad-based {Neural} {Network} for {Coreference} {Resolution}", language = "en", journal = "Proceedings of COLING", year = "2018", pages = "9" }