Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384644539> ?p ?o ?g. }
Showing items 1 to 89 of
89
with 100 items per page.
- W4384644539 abstract "Learning vectors that capture the meaning of concepts remains a fundamental challenge. Somewhat surprisingly, perhaps, pre-trained language models have thus far only enabled modest improvements to the quality of such concept embeddings. Current strategies for using language models typically represent a concept by averaging the contextualised representations of its mentions in some corpus. This is potentially sub-optimal for at least two reasons. First, contextualised word vectors have an unusual geometry, which hampers downstream tasks. Second, concept embeddings should capture the semantic properties of concepts, whereas contextualised word vectors are also affected by other factors. To address these issues, we propose two contrastive learning strategies, based on the view that whenever two sentences reveal similar properties, the corresponding contextualised vectors should also be similar. One strategy is fully unsupervised, estimating the properties which are expressed in a sentence from the neighbourhood structure of the contextualised word embeddings. The second strategy instead relies on a distant supervision signal from ConceptNet. Our experimental results show that the resulting vectors substantially outperform existing concept embeddings in predicting the semantic properties of concepts, with the ConceptNet-based strategy achieving the best results. These findings are furthermore confirmed in a clustering task and in the downstream task of ontology completion." @default.
- W4384644539 created "2023-07-19" @default.
- W4384644539 creator A5009887598 @default.
- W4384644539 creator A5028866522 @default.
- W4384644539 creator A5059764145 @default.
- W4384644539 creator A5091979023 @default.
- W4384644539 date "2023-07-18" @default.
- W4384644539 modified "2023-09-26" @default.
- W4384644539 title "Distilling Semantic Concept Embeddings from Contrastively Fine-Tuned Language Models" @default.
- W4384644539 cites W2050482109 @default.
- W4384644539 cites W2078894097 @default.
- W4384644539 cites W2167949727 @default.
- W4384644539 cites W2250539671 @default.
- W4384644539 cites W2251803266 @default.
- W4384644539 cites W2561529111 @default.
- W4384644539 cites W2740540529 @default.
- W4384644539 cites W2803524328 @default.
- W4384644539 cites W2882319491 @default.
- W4384644539 cites W2949952652 @default.
- W4384644539 cites W2951684477 @default.
- W4384644539 cites W2963158304 @default.
- W4384644539 cites W3035044482 @default.
- W4384644539 cites W3035102548 @default.
- W4384644539 cites W3045464143 @default.
- W4384644539 cites W3098614164 @default.
- W4384644539 cites W3099475240 @default.
- W4384644539 cites W3105816068 @default.
- W4384644539 cites W3112012747 @default.
- W4384644539 cites W3135970112 @default.
- W4384644539 cites W3156636935 @default.
- W4384644539 cites W3191562205 @default.
- W4384644539 cites W3196813608 @default.
- W4384644539 cites W3197599296 @default.
- W4384644539 cites W3200327212 @default.
- W4384644539 cites W3201193395 @default.
- W4384644539 cites W3213730158 @default.
- W4384644539 cites W3215828051 @default.
- W4384644539 cites W3217229393 @default.
- W4384644539 cites W4283819412 @default.
- W4384644539 doi "https://doi.org/10.1145/3539618.3591667" @default.
- W4384644539 hasPublicationYear "2023" @default.
- W4384644539 type Work @default.
- W4384644539 citedByCount "0" @default.
- W4384644539 crossrefType "proceedings-article" @default.
- W4384644539 hasAuthorship W4384644539A5009887598 @default.
- W4384644539 hasAuthorship W4384644539A5028866522 @default.
- W4384644539 hasAuthorship W4384644539A5059764145 @default.
- W4384644539 hasAuthorship W4384644539A5091979023 @default.
- W4384644539 hasBestOaLocation W43846445392 @default.
- W4384644539 hasConcept C138885662 @default.
- W4384644539 hasConcept C154945302 @default.
- W4384644539 hasConcept C162324750 @default.
- W4384644539 hasConcept C187736073 @default.
- W4384644539 hasConcept C204321447 @default.
- W4384644539 hasConcept C2777530160 @default.
- W4384644539 hasConcept C2780451532 @default.
- W4384644539 hasConcept C41008148 @default.
- W4384644539 hasConcept C41895202 @default.
- W4384644539 hasConcept C73555534 @default.
- W4384644539 hasConcept C90805587 @default.
- W4384644539 hasConceptScore W4384644539C138885662 @default.
- W4384644539 hasConceptScore W4384644539C154945302 @default.
- W4384644539 hasConceptScore W4384644539C162324750 @default.
- W4384644539 hasConceptScore W4384644539C187736073 @default.
- W4384644539 hasConceptScore W4384644539C204321447 @default.
- W4384644539 hasConceptScore W4384644539C2777530160 @default.
- W4384644539 hasConceptScore W4384644539C2780451532 @default.
- W4384644539 hasConceptScore W4384644539C41008148 @default.
- W4384644539 hasConceptScore W4384644539C41895202 @default.
- W4384644539 hasConceptScore W4384644539C73555534 @default.
- W4384644539 hasConceptScore W4384644539C90805587 @default.
- W4384644539 hasLocation W43846445391 @default.
- W4384644539 hasLocation W43846445392 @default.
- W4384644539 hasLocation W43846445393 @default.
- W4384644539 hasOpenAccess W4384644539 @default.
- W4384644539 hasPrimaryLocation W43846445391 @default.
- W4384644539 hasRelatedWork W1539050421 @default.
- W4384644539 hasRelatedWork W1563147278 @default.
- W4384644539 hasRelatedWork W1573537589 @default.
- W4384644539 hasRelatedWork W159132833 @default.
- W4384644539 hasRelatedWork W2125145484 @default.
- W4384644539 hasRelatedWork W2903246208 @default.
- W4384644539 hasRelatedWork W2969773591 @default.
- W4384644539 hasRelatedWork W3024381485 @default.
- W4384644539 hasRelatedWork W4225619937 @default.
- W4384644539 hasRelatedWork W4385572842 @default.
- W4384644539 isParatext "false" @default.
- W4384644539 isRetracted "false" @default.
- W4384644539 workType "article" @default.