Matches in SemOpenAlex for { <https://semopenalex.org/work/W4226307002> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W4226307002 abstract "In this paper, we approach the problem of semantic search by framing the search task as paraphrase span detection, i.e. given a segment of text as a query phrase, the task is to identify its paraphrase in a given document, the same modelling setup as typically used in extractive question answering. On the Turku Paraphrase Corpus of 100,000 manually extracted Finnish paraphrase pairs including their original document context, we find that our paraphrase span detection model outperforms two strong retrieval baselines (lexical similarity and BERT sentence embeddings) by 31.9pp and 22.4pp respectively in terms of exact match, and by 22.3pp and 12.9pp in terms of token-level F-score. This demonstrates a strong advantage of modelling the task in terms of span retrieval, rather than sentence similarity. Additionally, we introduce a method for creating artificial paraphrase data through back-translation, suitable for languages where manually annotated paraphrase resources for training the span detection model are not available." @default.
- W4226307002 created "2022-05-05" @default.
- W4226307002 creator A5006363078 @default.
- W4226307002 creator A5019929457 @default.
- W4226307002 creator A5048932608 @default.
- W4226307002 creator A5053401029 @default.
- W4226307002 creator A5061159001 @default.
- W4226307002 creator A5073565606 @default.
- W4226307002 date "2021-12-09" @default.
- W4226307002 modified "2023-09-29" @default.
- W4226307002 title "Semantic Search as Extractive Paraphrase Span Detection" @default.
- W4226307002 hasPublicationYear "2021" @default.
- W4226307002 type Work @default.
- W4226307002 citedByCount "0" @default.
- W4226307002 crossrefType "posted-content" @default.
- W4226307002 hasAuthorship W4226307002A5006363078 @default.
- W4226307002 hasAuthorship W4226307002A5019929457 @default.
- W4226307002 hasAuthorship W4226307002A5048932608 @default.
- W4226307002 hasAuthorship W4226307002A5053401029 @default.
- W4226307002 hasAuthorship W4226307002A5061159001 @default.
- W4226307002 hasAuthorship W4226307002A5073565606 @default.
- W4226307002 hasBestOaLocation W42263070021 @default.
- W4226307002 hasConcept C130318100 @default.
- W4226307002 hasConcept C151730666 @default.
- W4226307002 hasConcept C154945302 @default.
- W4226307002 hasConcept C162324750 @default.
- W4226307002 hasConcept C187736073 @default.
- W4226307002 hasConcept C203005215 @default.
- W4226307002 hasConcept C204321447 @default.
- W4226307002 hasConcept C23123220 @default.
- W4226307002 hasConcept C2776224158 @default.
- W4226307002 hasConcept C2777530160 @default.
- W4226307002 hasConcept C2779343474 @default.
- W4226307002 hasConcept C2780451532 @default.
- W4226307002 hasConcept C2780907237 @default.
- W4226307002 hasConcept C2780922921 @default.
- W4226307002 hasConcept C2985367798 @default.
- W4226307002 hasConcept C41008148 @default.
- W4226307002 hasConcept C44291984 @default.
- W4226307002 hasConcept C86803240 @default.
- W4226307002 hasConceptScore W4226307002C130318100 @default.
- W4226307002 hasConceptScore W4226307002C151730666 @default.
- W4226307002 hasConceptScore W4226307002C154945302 @default.
- W4226307002 hasConceptScore W4226307002C162324750 @default.
- W4226307002 hasConceptScore W4226307002C187736073 @default.
- W4226307002 hasConceptScore W4226307002C203005215 @default.
- W4226307002 hasConceptScore W4226307002C204321447 @default.
- W4226307002 hasConceptScore W4226307002C23123220 @default.
- W4226307002 hasConceptScore W4226307002C2776224158 @default.
- W4226307002 hasConceptScore W4226307002C2777530160 @default.
- W4226307002 hasConceptScore W4226307002C2779343474 @default.
- W4226307002 hasConceptScore W4226307002C2780451532 @default.
- W4226307002 hasConceptScore W4226307002C2780907237 @default.
- W4226307002 hasConceptScore W4226307002C2780922921 @default.
- W4226307002 hasConceptScore W4226307002C2985367798 @default.
- W4226307002 hasConceptScore W4226307002C41008148 @default.
- W4226307002 hasConceptScore W4226307002C44291984 @default.
- W4226307002 hasConceptScore W4226307002C86803240 @default.
- W4226307002 hasLocation W42263070021 @default.
- W4226307002 hasOpenAccess W4226307002 @default.
- W4226307002 hasPrimaryLocation W42263070021 @default.
- W4226307002 hasRelatedWork W13206174 @default.
- W4226307002 hasRelatedWork W14808 @default.
- W4226307002 hasRelatedWork W149980 @default.
- W4226307002 hasRelatedWork W1788602 @default.
- W4226307002 hasRelatedWork W2060686 @default.
- W4226307002 hasRelatedWork W505434 @default.
- W4226307002 hasRelatedWork W745926 @default.
- W4226307002 hasRelatedWork W881173 @default.
- W4226307002 hasRelatedWork W8895266 @default.
- W4226307002 hasRelatedWork W8411197 @default.
- W4226307002 isParatext "false" @default.
- W4226307002 isRetracted "false" @default.
- W4226307002 workType "article" @default.