We propose a local coherence model based on a convolutional neural network that operates over the entity grid representation of a text. The model captures long range entity transitions along with entity-specific features without loosing generalization, thanks to the power of distributed representation. We present a pairwise ranking method to train the model in an end-to-end fashion on a task and learn task-specific high level features. Our evaluation on three different coherence assessment tasks demonstrates that our model achieves state of the art results outperforming existing models by a good margin.

@inproceedings{nguyen-joty-acl-17, abstract = {We propose a local coherence model based on a convolutional neural network that operates over the entity grid representation of a text. The model captures long range entity transitions along with entity-specific features without loosing generalization, thanks to the power of distributed representation. We present a pairwise ranking method to train the model in an end-to-end fashion on a task and learn task-specific high level features. Our evaluation on three different coherence assessment tasks demonstrates that our model achieves state of the art results outperforming existing models by a good margin.}, address = {Vancouver, Canada}, author = {Dat Nguyen* and Shafiq Joty*}, booktitle = {Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics}, month = {August}, pages = {1320--1330}, publisher = {Association for Computational Linguistics}, series = {ACL'17}, title = {A Neural Local Coherence Model}, url = {papers/nguyen-joty-acl-17.pdf}, year = {2017} }