Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897527206> ?p ?o ?g. }
- W2897527206 abstract "Measuring the semantic similarity between two sentences (or Semantic Textual Similarity - STS) is fundamental in many NLP applications. Despite the remarkable results in supervised settings with adequate labeling, little attention has been paid to this task in low-resource languages with insufficient labeling. Existing approaches mostly leverage machine translation techniques to translate sentences into rich-resource language. These approaches either beget language biases, or be impractical in industrial applications where spoken language scenario is more often and rigorous efficiency is required. In this work, we propose a multilingual framework to tackle the STS task in a low-resource language e.g. Spanish, Arabic , Indonesian and Thai, by utilizing the rich annotation data in a rich resource language, e.g. English. Our approach is extended from a basic monolingual STS framework to a shared multilingual encoder pretrained with translation task to incorporate rich-resource language data. By exploiting the nature of a shared multilingual encoder, one sentence can have multiple representations for different target translation language, which are used in an ensemble model to improve similarity evaluation. We demonstrate the superiority of our method over other state of the art approaches on SemEval STS task by its significant improvement on non-MT method, as well as an online industrial product where MT method fails to beat baseline while our approach still has consistently improvements." @default.
- W2897527206 created "2018-10-26" @default.
- W2897527206 creator A5007613197 @default.
- W2897527206 creator A5009559278 @default.
- W2897527206 creator A5021428185 @default.
- W2897527206 creator A5029840191 @default.
- W2897527206 creator A5046413480 @default.
- W2897527206 creator A5048233257 @default.
- W2897527206 creator A5058025659 @default.
- W2897527206 creator A5071530717 @default.
- W2897527206 date "2018-10-20" @default.
- W2897527206 modified "2023-09-26" @default.
- W2897527206 title "Improving Multilingual Semantic Textual Similarity with Shared Sentence Encoder for Low-resource Languages" @default.
- W2897527206 cites W1816313093 @default.
- W2897527206 cites W2104246439 @default.
- W2897527206 cites W2145094598 @default.
- W2897527206 cites W2157331557 @default.
- W2897527206 cites W2211192759 @default.
- W2897527206 cites W2251427843 @default.
- W2897527206 cites W2280395961 @default.
- W2897527206 cites W2413794162 @default.
- W2897527206 cites W2493916176 @default.
- W2897527206 cites W2626778328 @default.
- W2897527206 cites W2740747242 @default.
- W2897527206 cites W2751762827 @default.
- W2897527206 cites W2752289368 @default.
- W2897527206 cites W2786685006 @default.
- W2897527206 cites W2787560479 @default.
- W2897527206 cites W2794557536 @default.
- W2897527206 cites W2949667497 @default.
- W2897527206 cites W2949888546 @default.
- W2897527206 cites W2949989304 @default.
- W2897527206 cites W2950133940 @default.
- W2897527206 cites W2962824887 @default.
- W2897527206 cites W2963661177 @default.
- W2897527206 cites W3104033643 @default.
- W2897527206 cites W630532510 @default.
- W2897527206 hasPublicationYear "2018" @default.
- W2897527206 type Work @default.
- W2897527206 sameAs 2897527206 @default.
- W2897527206 citedByCount "1" @default.
- W2897527206 countsByYear W28975272062019 @default.
- W2897527206 crossrefType "posted-content" @default.
- W2897527206 hasAuthorship W2897527206A5007613197 @default.
- W2897527206 hasAuthorship W2897527206A5009559278 @default.
- W2897527206 hasAuthorship W2897527206A5021428185 @default.
- W2897527206 hasAuthorship W2897527206A5029840191 @default.
- W2897527206 hasAuthorship W2897527206A5046413480 @default.
- W2897527206 hasAuthorship W2897527206A5048233257 @default.
- W2897527206 hasAuthorship W2897527206A5058025659 @default.
- W2897527206 hasAuthorship W2897527206A5071530717 @default.
- W2897527206 hasConcept C103278499 @default.
- W2897527206 hasConcept C111919701 @default.
- W2897527206 hasConcept C115961682 @default.
- W2897527206 hasConcept C118505674 @default.
- W2897527206 hasConcept C130318100 @default.
- W2897527206 hasConcept C153083717 @default.
- W2897527206 hasConcept C154945302 @default.
- W2897527206 hasConcept C162324750 @default.
- W2897527206 hasConcept C187736073 @default.
- W2897527206 hasConcept C203005215 @default.
- W2897527206 hasConcept C204321447 @default.
- W2897527206 hasConcept C206345919 @default.
- W2897527206 hasConcept C2777530160 @default.
- W2897527206 hasConcept C2780451532 @default.
- W2897527206 hasConcept C31258907 @default.
- W2897527206 hasConcept C41008148 @default.
- W2897527206 hasConceptScore W2897527206C103278499 @default.
- W2897527206 hasConceptScore W2897527206C111919701 @default.
- W2897527206 hasConceptScore W2897527206C115961682 @default.
- W2897527206 hasConceptScore W2897527206C118505674 @default.
- W2897527206 hasConceptScore W2897527206C130318100 @default.
- W2897527206 hasConceptScore W2897527206C153083717 @default.
- W2897527206 hasConceptScore W2897527206C154945302 @default.
- W2897527206 hasConceptScore W2897527206C162324750 @default.
- W2897527206 hasConceptScore W2897527206C187736073 @default.
- W2897527206 hasConceptScore W2897527206C203005215 @default.
- W2897527206 hasConceptScore W2897527206C204321447 @default.
- W2897527206 hasConceptScore W2897527206C206345919 @default.
- W2897527206 hasConceptScore W2897527206C2777530160 @default.
- W2897527206 hasConceptScore W2897527206C2780451532 @default.
- W2897527206 hasConceptScore W2897527206C31258907 @default.
- W2897527206 hasConceptScore W2897527206C41008148 @default.
- W2897527206 hasLocation W28975272061 @default.
- W2897527206 hasOpenAccess W2897527206 @default.
- W2897527206 hasPrimaryLocation W28975272061 @default.
- W2897527206 hasRelatedWork W2251678997 @default.
- W2897527206 hasRelatedWork W2252092358 @default.
- W2897527206 hasRelatedWork W2789718859 @default.
- W2897527206 hasRelatedWork W2791169651 @default.
- W2897527206 hasRelatedWork W2886711982 @default.
- W2897527206 hasRelatedWork W2889191148 @default.
- W2897527206 hasRelatedWork W2895414369 @default.
- W2897527206 hasRelatedWork W2923290299 @default.
- W2897527206 hasRelatedWork W2943490406 @default.
- W2897527206 hasRelatedWork W2947015347 @default.
- W2897527206 hasRelatedWork W2948526413 @default.
- W2897527206 hasRelatedWork W2952289666 @default.
- W2897527206 hasRelatedWork W2963863756 @default.
- W2897527206 hasRelatedWork W3097619042 @default.