Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387521241> ?p ?o ?g. }
Showing items 1 to 73 of
73
with 100 items per page.
- W4387521241 endingPage "121909" @default.
- W4387521241 startingPage "121909" @default.
- W4387521241 abstract "Sentence representation is one of the most fundamental research topics in natural language processing (NLP), as its quality directly affects various downstream task performances. Recent studies for sentence representations have established state-of-the-art (SOTA) performance on semantic representation tasks. However, embeddings by those approaches share unsatisfying transferability when applied to various specific applications. Seldom work studies the transferability of semantic sentence embeddings. In this paper, we first explore the transferability characteristic of the sentence embeddings, and present BlendCSE, a new sentence embedding model targeting rich semantics and transferability. BlendCSE blends three recent advanced NLP learning methodologies, namely, continue learning on masked language modeling (MLM), contrastive learning (CL) with data augmentations (DA), and semantic supervised learning. The main objectives of BlendCSE are to capture token/word level information, diversified linguistic properties, and sentence semantics, respectively. Empirical studies demonstrate that BlendCSE captures semantics comparably well on STS tasks, yet surpasses existing methods on various transfer tasks, yielding even stronger transferability on document-level applications. Ablation studies verified that the three learning objectives synergy well to capture semantics and transferability effectively." @default.
- W4387521241 created "2023-10-12" @default.
- W4387521241 creator A5027332330 @default.
- W4387521241 creator A5036993751 @default.
- W4387521241 creator A5084738263 @default.
- W4387521241 creator A5093041427 @default.
- W4387521241 date "2023-10-01" @default.
- W4387521241 modified "2023-10-12" @default.
- W4387521241 title "BlendCSE: Blend contrastive learnings for sentence embeddings with rich semantics and transferability" @default.
- W4387521241 cites W1493526108 @default.
- W4387521241 cites W1980776243 @default.
- W4387521241 cites W2014902591 @default.
- W4387521241 cites W2028175314 @default.
- W4387521241 cites W2114524997 @default.
- W4387521241 cites W2126400076 @default.
- W4387521241 cites W2133458109 @default.
- W4387521241 cites W2160660844 @default.
- W4387521241 cites W2163455955 @default.
- W4387521241 cites W2778443828 @default.
- W4387521241 cites W2891177506 @default.
- W4387521241 cites W2963918774 @default.
- W4387521241 cites W3035524453 @default.
- W4387521241 cites W3042631625 @default.
- W4387521241 cites W3105816068 @default.
- W4387521241 doi "https://doi.org/10.1016/j.eswa.2023.121909" @default.
- W4387521241 hasPublicationYear "2023" @default.
- W4387521241 type Work @default.
- W4387521241 citedByCount "0" @default.
- W4387521241 crossrefType "journal-article" @default.
- W4387521241 hasAuthorship W4387521241A5027332330 @default.
- W4387521241 hasAuthorship W4387521241A5036993751 @default.
- W4387521241 hasAuthorship W4387521241A5084738263 @default.
- W4387521241 hasAuthorship W4387521241A5093041427 @default.
- W4387521241 hasConcept C119857082 @default.
- W4387521241 hasConcept C140331021 @default.
- W4387521241 hasConcept C150899416 @default.
- W4387521241 hasConcept C154945302 @default.
- W4387521241 hasConcept C184337299 @default.
- W4387521241 hasConcept C199360897 @default.
- W4387521241 hasConcept C204321447 @default.
- W4387521241 hasConcept C2777530160 @default.
- W4387521241 hasConcept C41008148 @default.
- W4387521241 hasConcept C41608201 @default.
- W4387521241 hasConcept C61272859 @default.
- W4387521241 hasConceptScore W4387521241C119857082 @default.
- W4387521241 hasConceptScore W4387521241C140331021 @default.
- W4387521241 hasConceptScore W4387521241C150899416 @default.
- W4387521241 hasConceptScore W4387521241C154945302 @default.
- W4387521241 hasConceptScore W4387521241C184337299 @default.
- W4387521241 hasConceptScore W4387521241C199360897 @default.
- W4387521241 hasConceptScore W4387521241C204321447 @default.
- W4387521241 hasConceptScore W4387521241C2777530160 @default.
- W4387521241 hasConceptScore W4387521241C41008148 @default.
- W4387521241 hasConceptScore W4387521241C41608201 @default.
- W4387521241 hasConceptScore W4387521241C61272859 @default.
- W4387521241 hasLocation W43875212411 @default.
- W4387521241 hasOpenAccess W4387521241 @default.
- W4387521241 hasPrimaryLocation W43875212411 @default.
- W4387521241 hasRelatedWork W2011110943 @default.
- W4387521241 hasRelatedWork W2011433332 @default.
- W4387521241 hasRelatedWork W2055748329 @default.
- W4387521241 hasRelatedWork W2161221533 @default.
- W4387521241 hasRelatedWork W2216382288 @default.
- W4387521241 hasRelatedWork W2355491300 @default.
- W4387521241 hasRelatedWork W2582594227 @default.
- W4387521241 hasRelatedWork W4234629551 @default.
- W4387521241 hasRelatedWork W1534779234 @default.
- W4387521241 hasRelatedWork W1666484574 @default.
- W4387521241 isParatext "false" @default.
- W4387521241 isRetracted "false" @default.
- W4387521241 workType "article" @default.