Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378465250> ?p ?o ?g. }
Showing items 1 to 51 of
51
with 100 items per page.
- W4378465250 abstract "Pre-trained language models (PLMs) were considered to be able to store relational knowledge present in the training data. However, some relational knowledge seems to be discarded unsafely in PLMs due to textbf{report bias}: low-frequency relational knowledge might be underexpressed compared to high-frequency one in PLMs. This gives us a hint that relational knowledge might not be redundant to the stored knowledge of PLMs, but rather be complementary. To additionally inject relational knowledge into PLMs, we propose a simple-yet-effective approach to inject relational knowledge into PLMs, which is inspired by three observations (namely, polymorphism, synonymous substitution, and association). In particular, we switch entities in the training corpus to related entities (either hypernyms/hyponyms/synonyms, or arbitrarily-related concepts). Experimental results show that the proposed approach could not only better capture relational knowledge, but also improve the performance in various biomedical downstream tasks. Our model is available in url{https://github.com/StevenZHB/BioPLM_InjectingKnowledge}." @default.
- W4378465250 created "2023-05-27" @default.
- W4378465250 creator A5057282504 @default.
- W4378465250 creator A5064737783 @default.
- W4378465250 creator A5090567275 @default.
- W4378465250 date "2023-05-24" @default.
- W4378465250 modified "2023-09-25" @default.
- W4378465250 title "Injecting Knowledge into Biomedical Pre-trained Models via Polymorphism and Synonymous Substitution" @default.
- W4378465250 doi "https://doi.org/10.48550/arxiv.2305.15010" @default.
- W4378465250 hasPublicationYear "2023" @default.
- W4378465250 type Work @default.
- W4378465250 citedByCount "0" @default.
- W4378465250 crossrefType "posted-content" @default.
- W4378465250 hasAuthorship W4378465250A5057282504 @default.
- W4378465250 hasAuthorship W4378465250A5064737783 @default.
- W4378465250 hasAuthorship W4378465250A5090567275 @default.
- W4378465250 hasBestOaLocation W43784652501 @default.
- W4378465250 hasConcept C119857082 @default.
- W4378465250 hasConcept C154945302 @default.
- W4378465250 hasConcept C177877439 @default.
- W4378465250 hasConcept C199360897 @default.
- W4378465250 hasConcept C204321447 @default.
- W4378465250 hasConcept C23123220 @default.
- W4378465250 hasConcept C2778220771 @default.
- W4378465250 hasConcept C41008148 @default.
- W4378465250 hasConcept C5655090 @default.
- W4378465250 hasConceptScore W4378465250C119857082 @default.
- W4378465250 hasConceptScore W4378465250C154945302 @default.
- W4378465250 hasConceptScore W4378465250C177877439 @default.
- W4378465250 hasConceptScore W4378465250C199360897 @default.
- W4378465250 hasConceptScore W4378465250C204321447 @default.
- W4378465250 hasConceptScore W4378465250C23123220 @default.
- W4378465250 hasConceptScore W4378465250C2778220771 @default.
- W4378465250 hasConceptScore W4378465250C41008148 @default.
- W4378465250 hasConceptScore W4378465250C5655090 @default.
- W4378465250 hasLocation W43784652501 @default.
- W4378465250 hasOpenAccess W4378465250 @default.
- W4378465250 hasPrimaryLocation W43784652501 @default.
- W4378465250 hasRelatedWork W1585375882 @default.
- W4378465250 hasRelatedWork W2485866096 @default.
- W4378465250 hasRelatedWork W2611614995 @default.
- W4378465250 hasRelatedWork W2961085424 @default.
- W4378465250 hasRelatedWork W3095868780 @default.
- W4378465250 hasRelatedWork W3216470990 @default.
- W4378465250 hasRelatedWork W4286629047 @default.
- W4378465250 hasRelatedWork W4306321456 @default.
- W4378465250 hasRelatedWork W4306674287 @default.
- W4378465250 hasRelatedWork W4224009465 @default.
- W4378465250 isParatext "false" @default.
- W4378465250 isRetracted "false" @default.
- W4378465250 workType "article" @default.