Matches in SemOpenAlex for { <https://semopenalex.org/work/W4287777256> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4287777256 abstract "Simple weighted averaging of word vectors often yields effective representations for sentences which outperform sophisticated seq2seq neural models in many tasks. While it is desirable to use the same method to represent documents as well, unfortunately, the effectiveness is lost when representing long documents involving multiple sentences. One of the key reasons is that a longer document is likely to contain words from many different topics; hence, creating a single vector while ignoring all the topical structure is unlikely to yield an effective document representation. This problem is less acute in single sentences and other short text fragments where the presence of a single topic is most likely. To alleviate this problem, we present P-SIF, a partitioned word averaging model to represent long documents. P-SIF retains the simplicity of simple weighted word averaging while taking a document's topical structure into account. In particular, P-SIF learns topic-specific vectors from a document and finally concatenates them all to represent the overall document. We provide theoretical justifications on the correctness of P-SIF. Through a comprehensive set of experiments, we demonstrate P-SIF's effectiveness compared to simple weighted averaging and many other baselines." @default.
- W4287777256 created "2022-07-26" @default.
- W4287777256 creator A5007012613 @default.
- W4287777256 creator A5029651089 @default.
- W4287777256 creator A5031180836 @default.
- W4287777256 creator A5033387018 @default.
- W4287777256 creator A5033696194 @default.
- W4287777256 creator A5048002784 @default.
- W4287777256 date "2020-05-18" @default.
- W4287777256 modified "2023-09-30" @default.
- W4287777256 title "P-SIF: Document Embeddings Using Partition Averaging" @default.
- W4287777256 doi "https://doi.org/10.48550/arxiv.2005.09069" @default.
- W4287777256 hasPublicationYear "2020" @default.
- W4287777256 type Work @default.
- W4287777256 citedByCount "0" @default.
- W4287777256 crossrefType "posted-content" @default.
- W4287777256 hasAuthorship W4287777256A5007012613 @default.
- W4287777256 hasAuthorship W4287777256A5029651089 @default.
- W4287777256 hasAuthorship W4287777256A5031180836 @default.
- W4287777256 hasAuthorship W4287777256A5033387018 @default.
- W4287777256 hasAuthorship W4287777256A5033696194 @default.
- W4287777256 hasAuthorship W4287777256A5048002784 @default.
- W4287777256 hasBestOaLocation W42877772561 @default.
- W4287777256 hasConcept C111472728 @default.
- W4287777256 hasConcept C11413529 @default.
- W4287777256 hasConcept C114614502 @default.
- W4287777256 hasConcept C138885662 @default.
- W4287777256 hasConcept C153180895 @default.
- W4287777256 hasConcept C154945302 @default.
- W4287777256 hasConcept C177264268 @default.
- W4287777256 hasConcept C17744445 @default.
- W4287777256 hasConcept C199360897 @default.
- W4287777256 hasConcept C199539241 @default.
- W4287777256 hasConcept C204321447 @default.
- W4287777256 hasConcept C2524010 @default.
- W4287777256 hasConcept C2776359362 @default.
- W4287777256 hasConcept C2780586882 @default.
- W4287777256 hasConcept C33923547 @default.
- W4287777256 hasConcept C41008148 @default.
- W4287777256 hasConcept C42812 @default.
- W4287777256 hasConcept C55439883 @default.
- W4287777256 hasConcept C90805587 @default.
- W4287777256 hasConcept C94625758 @default.
- W4287777256 hasConceptScore W4287777256C111472728 @default.
- W4287777256 hasConceptScore W4287777256C11413529 @default.
- W4287777256 hasConceptScore W4287777256C114614502 @default.
- W4287777256 hasConceptScore W4287777256C138885662 @default.
- W4287777256 hasConceptScore W4287777256C153180895 @default.
- W4287777256 hasConceptScore W4287777256C154945302 @default.
- W4287777256 hasConceptScore W4287777256C177264268 @default.
- W4287777256 hasConceptScore W4287777256C17744445 @default.
- W4287777256 hasConceptScore W4287777256C199360897 @default.
- W4287777256 hasConceptScore W4287777256C199539241 @default.
- W4287777256 hasConceptScore W4287777256C204321447 @default.
- W4287777256 hasConceptScore W4287777256C2524010 @default.
- W4287777256 hasConceptScore W4287777256C2776359362 @default.
- W4287777256 hasConceptScore W4287777256C2780586882 @default.
- W4287777256 hasConceptScore W4287777256C33923547 @default.
- W4287777256 hasConceptScore W4287777256C41008148 @default.
- W4287777256 hasConceptScore W4287777256C42812 @default.
- W4287777256 hasConceptScore W4287777256C55439883 @default.
- W4287777256 hasConceptScore W4287777256C90805587 @default.
- W4287777256 hasConceptScore W4287777256C94625758 @default.
- W4287777256 hasLocation W42877772561 @default.
- W4287777256 hasLocation W42877772562 @default.
- W4287777256 hasOpenAccess W4287777256 @default.
- W4287777256 hasPrimaryLocation W42877772561 @default.
- W4287777256 hasRelatedWork W1517743118 @default.
- W4287777256 hasRelatedWork W2024218563 @default.
- W4287777256 hasRelatedWork W2072806201 @default.
- W4287777256 hasRelatedWork W2281090687 @default.
- W4287777256 hasRelatedWork W2952340579 @default.
- W4287777256 hasRelatedWork W2965845133 @default.
- W4287777256 hasRelatedWork W2985678088 @default.
- W4287777256 hasRelatedWork W4287629333 @default.
- W4287777256 hasRelatedWork W1602178951 @default.
- W4287777256 hasRelatedWork W1670831115 @default.
- W4287777256 isParatext "false" @default.
- W4287777256 isRetracted "false" @default.
- W4287777256 workType "article" @default.