Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385573332> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W4385573332 abstract "Text error correction aims to correct the errors in text sequences such as those typed by humans or generated by speech recognition models.Previous error correction methods usually take the source (incorrect) sentence as encoder input and generate the target (correct) sentence through the decoder. Since the error rate of the incorrect sentence is usually low (e.g., 10%), the correction model can only learn to correct on limited error tokens but trivially copy on most tokens (correct tokens), which harms the effective training of error correction. In this paper, we argue that the correct tokens should be better utilized to facilitate effective training and then propose a simple yet effective masking strategy to achieve this goal.Specifically, we randomly mask out a part of the correct tokens in the source sentence and let the model learn to not only correct the original error tokens but also predict the masked tokens based on their context information. Our method enjoys several advantages: 1) it alleviates trivial copy; 2) it leverages effective training signals from correct tokens; 3) it is a plug-and-play module and can be applied to different models and tasks. Experiments on spelling error correction and speech recognition error correction on Mandarin datasets and grammar error correction on English datasets with both autoregressive and non-autoregressive generation models show that our method improves the correctionaccuracy consistently." @default.
- W4385573332 created "2023-08-05" @default.
- W4385573332 creator A5005226552 @default.
- W4385573332 creator A5031767581 @default.
- W4385573332 creator A5034693159 @default.
- W4385573332 creator A5042906130 @default.
- W4385573332 creator A5053168506 @default.
- W4385573332 creator A5063062444 @default.
- W4385573332 creator A5068377419 @default.
- W4385573332 date "2022-01-01" @default.
- W4385573332 modified "2023-10-17" @default.
- W4385573332 title "Mask the Correct Tokens: An Embarrassingly Simple Approach for Error Correction" @default.
- W4385573332 doi "https://doi.org/10.18653/v1/2022.emnlp-main.708" @default.
- W4385573332 hasPublicationYear "2022" @default.
- W4385573332 type Work @default.
- W4385573332 citedByCount "1" @default.
- W4385573332 countsByYear W43855733322023 @default.
- W4385573332 crossrefType "proceedings-article" @default.
- W4385573332 hasAuthorship W4385573332A5005226552 @default.
- W4385573332 hasAuthorship W4385573332A5031767581 @default.
- W4385573332 hasAuthorship W4385573332A5034693159 @default.
- W4385573332 hasAuthorship W4385573332A5042906130 @default.
- W4385573332 hasAuthorship W4385573332A5053168506 @default.
- W4385573332 hasAuthorship W4385573332A5063062444 @default.
- W4385573332 hasAuthorship W4385573332A5068377419 @default.
- W4385573332 hasBestOaLocation W43855733321 @default.
- W4385573332 hasConcept C103088060 @default.
- W4385573332 hasConcept C11413529 @default.
- W4385573332 hasConcept C137293760 @default.
- W4385573332 hasConcept C151730666 @default.
- W4385573332 hasConcept C154945302 @default.
- W4385573332 hasConcept C204321447 @default.
- W4385573332 hasConcept C2777530160 @default.
- W4385573332 hasConcept C2779343474 @default.
- W4385573332 hasConcept C28490314 @default.
- W4385573332 hasConcept C40969351 @default.
- W4385573332 hasConcept C41008148 @default.
- W4385573332 hasConcept C86803240 @default.
- W4385573332 hasConceptScore W4385573332C103088060 @default.
- W4385573332 hasConceptScore W4385573332C11413529 @default.
- W4385573332 hasConceptScore W4385573332C137293760 @default.
- W4385573332 hasConceptScore W4385573332C151730666 @default.
- W4385573332 hasConceptScore W4385573332C154945302 @default.
- W4385573332 hasConceptScore W4385573332C204321447 @default.
- W4385573332 hasConceptScore W4385573332C2777530160 @default.
- W4385573332 hasConceptScore W4385573332C2779343474 @default.
- W4385573332 hasConceptScore W4385573332C28490314 @default.
- W4385573332 hasConceptScore W4385573332C40969351 @default.
- W4385573332 hasConceptScore W4385573332C41008148 @default.
- W4385573332 hasConceptScore W4385573332C86803240 @default.
- W4385573332 hasLocation W43855733321 @default.
- W4385573332 hasLocation W43855733322 @default.
- W4385573332 hasOpenAccess W4385573332 @default.
- W4385573332 hasPrimaryLocation W43855733321 @default.
- W4385573332 hasRelatedWork W1546181631 @default.
- W4385573332 hasRelatedWork W159132833 @default.
- W4385573332 hasRelatedWork W1989705153 @default.
- W4385573332 hasRelatedWork W2047094413 @default.
- W4385573332 hasRelatedWork W2080458333 @default.
- W4385573332 hasRelatedWork W20999564 @default.
- W4385573332 hasRelatedWork W2800921391 @default.
- W4385573332 hasRelatedWork W3024381485 @default.
- W4385573332 hasRelatedWork W3080136773 @default.
- W4385573332 hasRelatedWork W3171566221 @default.
- W4385573332 isParatext "false" @default.
- W4385573332 isRetracted "false" @default.
- W4385573332 workType "article" @default.