Please use this identifier to cite or link to this item: doi:10.22028/D291-30791
Volltext verfügbar? / Dokumentlieferung
Title: Next Sentence Prediction helps Implicit Discourse Relation Classification within and across Domains
Author(s): Shi, Wei
Demberg, Vera
Editor(s): Inui, Kentaro
Language: English
Title: 2019 Conference on Empirical Methods in Natural Language Processing and 9th International Joint Conference on Natural Language Processing - proceedings of the conference
Startpage: 5790
Endpage: 5796
Publisher/Platform: ACL
Year of Publication: 2019
Place of publication: Stroudsburg, PA
Title of the Conference: EMNLP-IJCNLP 2019
Place of the conference: Hong Kong, China
Publikation type: Conference Paper
Abstract: Implicit discourse relation classification is one of the most difficult tasks in discourse parsing. Previous studies have generally focused on extracting better representations of the relational arguments. In order to solve the task, it is however additionally necessary to capture what events are expected to cause or follow each other. Current discourse relation classifiers fall short in this respect. We here show that this shortcoming can be effectively addressed by using the bidirectional encoder representation from transformers (BERT) proposed by Devlin et al. (2019), which were trained on a next-sentence prediction task, and thus encode a representation of likely next sentences. The BERT-based model outperforms the current state of the art in 11-way classification by 8% points on the standard PDTB dataset. Our experiments also demonstrate that the model can be successfully ported to other domains: on the BioDRB dataset, the model outperforms the state of the art system around 15% points.
DOI of the first publication: 10.18653/v1/D19-1586
URL of the first publication: https://www.aclweb.org/anthology/D19-1586/
Link to this record: hdl:20.500.11880/29705
http://dx.doi.org/10.22028/D291-30791
ISBN: 978-1-950737-90-1
Date of registration: 23-Sep-2020
Faculty: MI - Fakultät für Mathematik und Informatik
Department: MI - Informatik
Professorship: MI - Prof. Dr. Vera Demberg
Collections:UniBib – Die Universitätsbibliographie

Files for this record:
There are no files associated with this item.


Items in SciDok are protected by copyright, with all rights reserved, unless otherwise indicated.