Matches in SemOpenAlex for { <https://semopenalex.org/work/W3201757274> ?p ?o ?g. }
Showing items 1 to 100 of
100
with 100 items per page.
- W3201757274 endingPage "584" @default.
- W3201757274 startingPage "570" @default.
- W3201757274 abstract "Encoder-decoder models have made great progress on handwritten mathematical expression recognition recently. However, it is still a challenge for existing methods to assign attention to image features accurately. Moreover, those encoder-decoder models usually adopt RNN-based models in their decoder part, which makes them inefficient in processing long Open image in new window sequences. In this paper, a transformer-based decoder is employed to replace RNN-based ones, which makes the whole model architecture very concise. Furthermore, a novel training strategy is introduced to fully exploit the potential of the transformer in bidirectional language modeling. Compared to several methods that do not use data augmentation, experiments demonstrate that our model improves the ExpRate of current state-of-the-art methods on CROHME 2014 by 2.23%. Similarly, on CROHME 2016 and CROHME 2019, we improve the ExpRate by 1.92% and 2.28% respectively." @default.
- W3201757274 created "2021-10-11" @default.
- W3201757274 creator A5004474713 @default.
- W3201757274 creator A5022625306 @default.
- W3201757274 creator A5025131129 @default.
- W3201757274 creator A5039730001 @default.
- W3201757274 creator A5047456533 @default.
- W3201757274 creator A5061781593 @default.
- W3201757274 date "2021-01-01" @default.
- W3201757274 modified "2023-10-16" @default.
- W3201757274 title "Handwritten Mathematical Expression Recognition with Bidirectionally Trained Transformer" @default.
- W3201757274 cites W1978799108 @default.
- W3201757274 cites W1998768285 @default.
- W3201757274 cites W2005079280 @default.
- W3201757274 cites W2031071334 @default.
- W3201757274 cites W2037121285 @default.
- W3201757274 cites W2072449446 @default.
- W3201757274 cites W2113029741 @default.
- W3201757274 cites W2194775991 @default.
- W3201757274 cites W2409722449 @default.
- W3201757274 cites W2466062786 @default.
- W3201757274 cites W2528252828 @default.
- W3201757274 cites W2623860192 @default.
- W3201757274 cites W2750938222 @default.
- W3201757274 cites W2785966045 @default.
- W3201757274 cites W2806199895 @default.
- W3201757274 cites W2886776719 @default.
- W3201757274 cites W2895924502 @default.
- W3201757274 cites W2935811960 @default.
- W3201757274 cites W2963446712 @default.
- W3201757274 cites W2963637207 @default.
- W3201757274 cites W2996840337 @default.
- W3201757274 cites W3004196288 @default.
- W3201757274 cites W3034644216 @default.
- W3201757274 cites W3110178865 @default.
- W3201757274 cites W3159162875 @default.
- W3201757274 cites W3161998573 @default.
- W3201757274 doi "https://doi.org/10.1007/978-3-030-86331-9_37" @default.
- W3201757274 hasPublicationYear "2021" @default.
- W3201757274 type Work @default.
- W3201757274 sameAs 3201757274 @default.
- W3201757274 citedByCount "25" @default.
- W3201757274 countsByYear W32017572742022 @default.
- W3201757274 countsByYear W32017572742023 @default.
- W3201757274 crossrefType "book-chapter" @default.
- W3201757274 hasAuthorship W3201757274A5004474713 @default.
- W3201757274 hasAuthorship W3201757274A5022625306 @default.
- W3201757274 hasAuthorship W3201757274A5025131129 @default.
- W3201757274 hasAuthorship W3201757274A5039730001 @default.
- W3201757274 hasAuthorship W3201757274A5047456533 @default.
- W3201757274 hasAuthorship W3201757274A5061781593 @default.
- W3201757274 hasBestOaLocation W32017572742 @default.
- W3201757274 hasConcept C111919701 @default.
- W3201757274 hasConcept C11413529 @default.
- W3201757274 hasConcept C118505674 @default.
- W3201757274 hasConcept C121332964 @default.
- W3201757274 hasConcept C153180895 @default.
- W3201757274 hasConcept C154945302 @default.
- W3201757274 hasConcept C165696696 @default.
- W3201757274 hasConcept C165801399 @default.
- W3201757274 hasConcept C28490314 @default.
- W3201757274 hasConcept C38652104 @default.
- W3201757274 hasConcept C41008148 @default.
- W3201757274 hasConcept C57273362 @default.
- W3201757274 hasConcept C62520636 @default.
- W3201757274 hasConcept C66322947 @default.
- W3201757274 hasConceptScore W3201757274C111919701 @default.
- W3201757274 hasConceptScore W3201757274C11413529 @default.
- W3201757274 hasConceptScore W3201757274C118505674 @default.
- W3201757274 hasConceptScore W3201757274C121332964 @default.
- W3201757274 hasConceptScore W3201757274C153180895 @default.
- W3201757274 hasConceptScore W3201757274C154945302 @default.
- W3201757274 hasConceptScore W3201757274C165696696 @default.
- W3201757274 hasConceptScore W3201757274C165801399 @default.
- W3201757274 hasConceptScore W3201757274C28490314 @default.
- W3201757274 hasConceptScore W3201757274C38652104 @default.
- W3201757274 hasConceptScore W3201757274C41008148 @default.
- W3201757274 hasConceptScore W3201757274C57273362 @default.
- W3201757274 hasConceptScore W3201757274C62520636 @default.
- W3201757274 hasConceptScore W3201757274C66322947 @default.
- W3201757274 hasLocation W32017572741 @default.
- W3201757274 hasLocation W32017572742 @default.
- W3201757274 hasOpenAccess W3201757274 @default.
- W3201757274 hasPrimaryLocation W32017572741 @default.
- W3201757274 hasRelatedWork W1496222301 @default.
- W3201757274 hasRelatedWork W1590307681 @default.
- W3201757274 hasRelatedWork W2296488620 @default.
- W3201757274 hasRelatedWork W2353836703 @default.
- W3201757274 hasRelatedWork W2358353312 @default.
- W3201757274 hasRelatedWork W3207760230 @default.
- W3201757274 hasRelatedWork W41015297 @default.
- W3201757274 hasRelatedWork W4280645561 @default.
- W3201757274 hasRelatedWork W4285370786 @default.
- W3201757274 hasRelatedWork W4312814274 @default.
- W3201757274 isParatext "false" @default.
- W3201757274 isRetracted "false" @default.
- W3201757274 magId "3201757274" @default.
- W3201757274 workType "book-chapter" @default.