This is the code of "Adversarial Connective-exploiting Networks for Implicit Discourse Relation Classification" in ACL 2017

Abstract:

Implicit discourse relation classification is of great challenge due to the lack of con- nectives as strong linguistic cues, which motivates the use of annotated implicit connectives to improve the recognition. We propose a feature imitation frame- work in which an implicit relation net- work is driven to learn from another neu- ral network with access to connectives, and thus encouraged to extract similarly salient features for accurate classification. We develop an adversarial model to en- able an adaptive imitation scheme through competition between the implicit network and a rival feature discriminator. Our method effectively transfers discriminabil- ity of connectives to the implicit features, and achieves state-of-the-art performance on the PDTB benchmark.

URL of this paper: http://www.aclweb.org/anthology/P/P17/P17-1093.pdf