Please use this identifier to cite or link to this item:
PIRA download icon_1.1View/Download Full Text
DC FieldValueLanguage
dc.contributorDepartment of Chinese and Bilingual Studiesen_US
dc.creatorChersoni, Een_US
dc.creatorPannitto, Len_US
dc.creatorSantus, Een_US
dc.creatorLenci, Aen_US
dc.creatorHuang, CRen_US
dc.rights© European Language Resources Association (ELRA), licensed under CC-BY-NC (
dc.rightsThe following publication Chersoni, E., Pannitto, L., Santus, E., Lenci, A., & Huang, C. R. (2020, May). Are Word Embeddings Really a Bad Fit for the Estimation of Thematic Fit?. In Proceedings of The 12th Language Resources and Evaluation Conference (pp. 5708-5713) is available at
dc.subjectCognitive methodsen_US
dc.subjectStatistical and machine learning methodsen_US
dc.titleAre word embeddings really a bad fit for the estimation of thematic fit?en_US
dc.typeConference Paperen_US
dcterms.abstractWhile neural embeddings represent a popular choice for word representation in a wide variety of NLP tasks, their usage for thematic fit modeling has been limited, as they have been reported to lag behind syntax-based count models. In this paper, we propose a complete evaluation of count models and word embeddings on thematic fit estimation, by taking into account a larger number of parameters and verb roles and introducing also dependency-based embeddings in the comparison. Our results show a complex scenario, where a determinant factor for the performance seems to be the availability to the model of reliable syntactic information for building the distributional representations of the roles.en_US
dcterms.accessRightsopen accessen_US
dcterms.bibliographicCitationProceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), Marseille, France, May 2020, p. 5708-5713en_US
dc.relation.ispartofbookProceedings of the 12th Language Resources and Evaluation Conferenceen_US
dc.relation.conferenceLanguage Resources and Evaluation Conferenceen_US
dc.description.validate202106 bcvcen_US
dc.description.oaVersion of Recorden_US
Appears in Collections:Conference Paper
Files in This Item:
File Description SizeFormat 
2020.lrec-1.700.pdf711.42 kBAdobe PDFView/Open
Open Access Information
Status open access
File Version Version of Record
View full-text via PolyU eLinks SFX Query
Show simple item record

Page views

Citations as of Jul 3, 2022


Citations as of Jul 3, 2022

Google ScholarTM


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