Show simple item record

dc.contributor.authorVulic, Ivan
dc.contributor.authorKorhonen, Anna-Leena
dc.contributor.authorLinguist, Assoc Computat
dc.date.accessioned2018-11-24T00:30:46Z
dc.date.available2018-11-24T00:30:46Z
dc.date.issued2018
dc.identifier.urihttps://www.repository.cam.ac.uk/handle/1810/285921
dc.description.abstractWord vector space specialisation models offer a portable, light-weight approach to fine-tuning arbitrary distributional vector spaces to discern between synonymy and antonymy. Their effectiveness is drawn from external linguistic constraints that specify the exact lexical relation between words. In this work, we show that a careful selection of the external constraints can steer and improve the specialisation. By simply selecting appropriate constraints, we report state-of-the-art results on a suite of tasks with well-defined benchmarks where modeling lexical contrast is crucial: 1) true semantic similarity, with highest reported scores on SimLex-999 and SimVerb-3500 to date; 2) detecting antonyms; and 3) distinguishing antonyms from synonyms.
dc.titleInjecting Lexical Contrast into Word Vectors by Guiding Vector Space Specialisation
dc.typeConference Object
prism.endingPage143
prism.publicationDate2018
prism.publicationNameREPRESENTATION LEARNING FOR NLP
prism.startingPage137
dc.identifier.doi10.17863/CAM.33248
dcterms.dateAccepted2018-05-18
rioxxterms.versionofrecord10.17863/CAM.33248
rioxxterms.licenseref.urihttp://www.rioxx.net/licenses/all-rights-reserved
rioxxterms.licenseref.startdate2018
rioxxterms.typeConference Paper/Proceeding/Abstract
pubs.funder-project-idEuropean Research Council (648909)
pubs.conference-nameProceedings of the 3rd Workshop on Representation Learning for NLP
pubs.conference-start-date2018-07-20
pubs.conference-finish-date2018-07-20
rioxxterms.freetoread.startdate2019-07-15


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record