Matches in SemOpenAlex for { <https://semopenalex.org/work/W4288094683> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4288094683 abstract "Similarity measures based purely on word embeddings are comfortably competing with much more sophisticated deep learning and expert-engineered systems on unsupervised semantic textual similarity (STS) tasks. In contrast to commonly used geometric approaches, we treat a single word embedding as e.g. 300 observations from a scalar random variable. Using this paradigm, we first illustrate that similarities derived from elementary pooling operations and classic correlation coefficients yield excellent results on standard STS benchmarks, outperforming many recently proposed methods while being much faster and trivial to implement. Next, we demonstrate how to avoid pooling operations altogether and compare sets of word embeddings directly via correlation operators between reproducing kernel Hilbert spaces. Just like cosine similarity is used to compare individual word vectors, we introduce a novel application of the centered kernel alignment (CKA) as a natural generalisation of squared cosine similarity for sets of word vectors. Likewise, CKA is very easy to implement and enjoys very strong empirical results." @default.
- W4288094683 created "2022-07-28" @default.
- W4288094683 creator A5019018683 @default.
- W4288094683 creator A5036590694 @default.
- W4288094683 creator A5044899440 @default.
- W4288094683 creator A5050567760 @default.
- W4288094683 creator A5067758665 @default.
- W4288094683 date "2019-10-07" @default.
- W4288094683 modified "2023-09-26" @default.
- W4288094683 title "Correlations between Word Vector Sets" @default.
- W4288094683 doi "https://doi.org/10.48550/arxiv.1910.02902" @default.
- W4288094683 hasPublicationYear "2019" @default.
- W4288094683 type Work @default.
- W4288094683 citedByCount "0" @default.
- W4288094683 crossrefType "posted-content" @default.
- W4288094683 hasAuthorship W4288094683A5019018683 @default.
- W4288094683 hasAuthorship W4288094683A5036590694 @default.
- W4288094683 hasAuthorship W4288094683A5044899440 @default.
- W4288094683 hasAuthorship W4288094683A5050567760 @default.
- W4288094683 hasAuthorship W4288094683A5067758665 @default.
- W4288094683 hasBestOaLocation W42880946831 @default.
- W4288094683 hasConcept C103278499 @default.
- W4288094683 hasConcept C11413529 @default.
- W4288094683 hasConcept C115961682 @default.
- W4288094683 hasConcept C118615104 @default.
- W4288094683 hasConcept C153180895 @default.
- W4288094683 hasConcept C154945302 @default.
- W4288094683 hasConcept C204321447 @default.
- W4288094683 hasConcept C2524010 @default.
- W4288094683 hasConcept C2777462759 @default.
- W4288094683 hasConcept C2780762811 @default.
- W4288094683 hasConcept C33923547 @default.
- W4288094683 hasConcept C41008148 @default.
- W4288094683 hasConcept C41608201 @default.
- W4288094683 hasConcept C70437156 @default.
- W4288094683 hasConcept C74193536 @default.
- W4288094683 hasConcept C80444323 @default.
- W4288094683 hasConcept C90805587 @default.
- W4288094683 hasConceptScore W4288094683C103278499 @default.
- W4288094683 hasConceptScore W4288094683C11413529 @default.
- W4288094683 hasConceptScore W4288094683C115961682 @default.
- W4288094683 hasConceptScore W4288094683C118615104 @default.
- W4288094683 hasConceptScore W4288094683C153180895 @default.
- W4288094683 hasConceptScore W4288094683C154945302 @default.
- W4288094683 hasConceptScore W4288094683C204321447 @default.
- W4288094683 hasConceptScore W4288094683C2524010 @default.
- W4288094683 hasConceptScore W4288094683C2777462759 @default.
- W4288094683 hasConceptScore W4288094683C2780762811 @default.
- W4288094683 hasConceptScore W4288094683C33923547 @default.
- W4288094683 hasConceptScore W4288094683C41008148 @default.
- W4288094683 hasConceptScore W4288094683C41608201 @default.
- W4288094683 hasConceptScore W4288094683C70437156 @default.
- W4288094683 hasConceptScore W4288094683C74193536 @default.
- W4288094683 hasConceptScore W4288094683C80444323 @default.
- W4288094683 hasConceptScore W4288094683C90805587 @default.
- W4288094683 hasLocation W42880946831 @default.
- W4288094683 hasOpenAccess W4288094683 @default.
- W4288094683 hasPrimaryLocation W42880946831 @default.
- W4288094683 hasRelatedWork W2264869955 @default.
- W4288094683 hasRelatedWork W2619127353 @default.
- W4288094683 hasRelatedWork W2921972128 @default.
- W4288094683 hasRelatedWork W2940384909 @default.
- W4288094683 hasRelatedWork W2949523605 @default.
- W4288094683 hasRelatedWork W2989727453 @default.
- W4288094683 hasRelatedWork W3029668006 @default.
- W4288094683 hasRelatedWork W3083244731 @default.
- W4288094683 hasRelatedWork W3113012686 @default.
- W4288094683 hasRelatedWork W3202766982 @default.
- W4288094683 isParatext "false" @default.
- W4288094683 isRetracted "false" @default.
- W4288094683 workType "article" @default.