Matches in SemOpenAlex for { <https://semopenalex.org/work/W4382404920> ?p ?o ?g. }
- W4382404920 endingPage "104431" @default.
- W4382404920 startingPage "104431" @default.
- W4382404920 abstract "In the era of digital healthcare, the huge volumes of textual information generated every day in hospitals constitute an essential but underused asset that could be exploited with task-specific, fine-tuned biomedical language representation models, improving patient care and management. For such specialized domains, previous research has shown that fine-tuning models stemming from broad-coverage checkpoints can largely benefit additional training rounds over large-scale in-domain resources. However, these resources are often unreachable for less-resourced languages like Italian, preventing local medical institutions to employ in-domain adaptation. In order to reduce this gap, our work investigates two accessible approaches to derive biomedical language models in languages other than English, taking Italian as a concrete use-case: one based on neural machine translation of English resources, favoring quantity over quality; the other based on a high-grade, narrow-scoped corpus natively written in Italian, thus preferring quality over quantity. Our study shows that data quantity is a harder constraint than data quality for biomedical adaptation, but the concatenation of high-quality data can improve model performance even when dealing with relatively size-limited corpora. The models published from our investigations have the potential to unlock important research opportunities for Italian hospitals and academia. Finally, the set of lessons learned from the study constitutes valuable insights towards a solution to build biomedical language models that are generalizable to other less-resourced languages and different domain settings." @default.
- W4382404920 created "2023-06-29" @default.
- W4382404920 creator A5046058191 @default.
- W4382404920 creator A5050932220 @default.
- W4382404920 creator A5073183618 @default.
- W4382404920 creator A5084685551 @default.
- W4382404920 creator A5090028219 @default.
- W4382404920 date "2023-08-01" @default.
- W4382404920 modified "2023-10-16" @default.
- W4382404920 title "Localizing in-domain adaptation of transformer-based biomedical language models" @default.
- W4382404920 cites W1981208470 @default.
- W4382404920 cites W2071879021 @default.
- W4382404920 cites W2132267839 @default.
- W4382404920 cites W2141559645 @default.
- W4382404920 cites W2149369282 @default.
- W4382404920 cites W2154142897 @default.
- W4382404920 cites W2169099542 @default.
- W4382404920 cites W2346452181 @default.
- W4382404920 cites W2396881363 @default.
- W4382404920 cites W2560647685 @default.
- W4382404920 cites W2768488789 @default.
- W4382404920 cites W2911489562 @default.
- W4382404920 cites W2965390203 @default.
- W4382404920 cites W3023545062 @default.
- W4382404920 cites W3199422761 @default.
- W4382404920 cites W4283318325 @default.
- W4382404920 cites W4285807172 @default.
- W4382404920 doi "https://doi.org/10.1016/j.jbi.2023.104431" @default.
- W4382404920 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37385327" @default.
- W4382404920 hasPublicationYear "2023" @default.
- W4382404920 type Work @default.
- W4382404920 citedByCount "0" @default.
- W4382404920 crossrefType "journal-article" @default.
- W4382404920 hasAuthorship W4382404920A5046058191 @default.
- W4382404920 hasAuthorship W4382404920A5050932220 @default.
- W4382404920 hasAuthorship W4382404920A5073183618 @default.
- W4382404920 hasAuthorship W4382404920A5084685551 @default.
- W4382404920 hasAuthorship W4382404920A5090028219 @default.
- W4382404920 hasBestOaLocation W43824049201 @default.
- W4382404920 hasConcept C111472728 @default.
- W4382404920 hasConcept C114614502 @default.
- W4382404920 hasConcept C120665830 @default.
- W4382404920 hasConcept C121332964 @default.
- W4382404920 hasConcept C134306372 @default.
- W4382404920 hasConcept C137293760 @default.
- W4382404920 hasConcept C138885662 @default.
- W4382404920 hasConcept C139807058 @default.
- W4382404920 hasConcept C154945302 @default.
- W4382404920 hasConcept C165801399 @default.
- W4382404920 hasConcept C203005215 @default.
- W4382404920 hasConcept C204321447 @default.
- W4382404920 hasConcept C2522767166 @default.
- W4382404920 hasConcept C2779530757 @default.
- W4382404920 hasConcept C33923547 @default.
- W4382404920 hasConcept C36503486 @default.
- W4382404920 hasConcept C41008148 @default.
- W4382404920 hasConcept C62520636 @default.
- W4382404920 hasConcept C66322947 @default.
- W4382404920 hasConcept C87619178 @default.
- W4382404920 hasConceptScore W4382404920C111472728 @default.
- W4382404920 hasConceptScore W4382404920C114614502 @default.
- W4382404920 hasConceptScore W4382404920C120665830 @default.
- W4382404920 hasConceptScore W4382404920C121332964 @default.
- W4382404920 hasConceptScore W4382404920C134306372 @default.
- W4382404920 hasConceptScore W4382404920C137293760 @default.
- W4382404920 hasConceptScore W4382404920C138885662 @default.
- W4382404920 hasConceptScore W4382404920C139807058 @default.
- W4382404920 hasConceptScore W4382404920C154945302 @default.
- W4382404920 hasConceptScore W4382404920C165801399 @default.
- W4382404920 hasConceptScore W4382404920C203005215 @default.
- W4382404920 hasConceptScore W4382404920C204321447 @default.
- W4382404920 hasConceptScore W4382404920C2522767166 @default.
- W4382404920 hasConceptScore W4382404920C2779530757 @default.
- W4382404920 hasConceptScore W4382404920C33923547 @default.
- W4382404920 hasConceptScore W4382404920C36503486 @default.
- W4382404920 hasConceptScore W4382404920C41008148 @default.
- W4382404920 hasConceptScore W4382404920C62520636 @default.
- W4382404920 hasConceptScore W4382404920C66322947 @default.
- W4382404920 hasConceptScore W4382404920C87619178 @default.
- W4382404920 hasFunder F4320320300 @default.
- W4382404920 hasFunder F4320321781 @default.
- W4382404920 hasFunder F4320327844 @default.
- W4382404920 hasLocation W43824049201 @default.
- W4382404920 hasLocation W43824049202 @default.
- W4382404920 hasLocation W43824049203 @default.
- W4382404920 hasOpenAccess W4382404920 @default.
- W4382404920 hasPrimaryLocation W43824049201 @default.
- W4382404920 hasRelatedWork W2888520903 @default.
- W4382404920 hasRelatedWork W2890256614 @default.
- W4382404920 hasRelatedWork W2903810591 @default.
- W4382404920 hasRelatedWork W2949454572 @default.
- W4382404920 hasRelatedWork W2963499882 @default.
- W4382404920 hasRelatedWork W2970247882 @default.
- W4382404920 hasRelatedWork W3026554633 @default.
- W4382404920 hasRelatedWork W3066373881 @default.
- W4382404920 hasRelatedWork W3098873988 @default.
- W4382404920 hasRelatedWork W3176018525 @default.
- W4382404920 hasVolume "144" @default.