FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3103818060> ?p ?o ?g. }

• W3103818060 abstract "Existing approaches for grammatical error correction (GEC) largely rely on supervised learning with manually created GEC datasets. However, there has been little focus on verifying and ensuring the quality of the datasets, and on how lower-quality data might affect GEC performance. We indeed found that there is a non-negligible amount of “noise” where errors were inappropriately edited or left uncorrected. To address this, we designed a self-refinement method where the key idea is to denoise these datasets by leveraging the prediction consistency of existing models, and outperformed strong denoising baseline methods. We further applied task-specific techniques and achieved state-of-the-art performance on the CoNLL-2014, JFLEG, and BEA-2019 benchmarks. We then analyzed the effect of the proposed denoising method, and found that our approach leads to improved coverage of corrections and facilitated fluency edits which are reflected in higher recall and overall performance." @default.
• W3103818060 created "2020-11-23" @default.
• W3103818060 creator A5002182453 @default.
• W3103818060 creator A5003760956 @default.
• W3103818060 creator A5005531754 @default.
• W3103818060 creator A5060910700 @default.
• W3103818060 creator A5066002127 @default.
• W3103818060 date "2020-01-01" @default.
• W3103818060 modified "2023-09-25" @default.
• W3103818060 title "A Self-Refinement Strategy for Noise Reduction in Grammatical Error Correction" @default.
• W3103818060 cites W1721115786 @default.
• W3103818060 cites W1996161790 @default.
• W3103818060 cites W2061693765 @default.
• W3103818060 cites W2123388068 @default.
• W3103818060 cites W2124725212 @default.
• W3103818060 cites W2125616599 @default.
• W3103818060 cites W2130942839 @default.
• W3103818060 cites W2144950812 @default.
• W3103818060 cites W2148474294 @default.
• W3103818060 cites W2153013403 @default.
• W3103818060 cites W2170527467 @default.
• W3103818060 cites W2183341477 @default.
• W3103818060 cites W2315316408 @default.
• W3103818060 cites W2375022080 @default.
• W3103818060 cites W2589277916 @default.
• W3103818060 cites W2740433069 @default.
• W3103818060 cites W2764337833 @default.
• W3103818060 cites W2797885244 @default.
• W3103818060 cites W2810035278 @default.
• W3103818060 cites W2889112299 @default.
• W3103818060 cites W2902319873 @default.
• W3103818060 cites W2902918014 @default.
• W3103818060 cites W2903490366 @default.
• W3103818060 cites W2931749839 @default.
• W3103818060 cites W2933138175 @default.
• W3103818060 cites W2936597270 @default.
• W3103818060 cites W2948335087 @default.
• W3103818060 cites W2949973181 @default.
• W3103818060 cites W2953035981 @default.
• W3103818060 cites W2962784628 @default.
• W3103818060 cites W2962801832 @default.
• W3103818060 cites W2963281280 @default.
• W3103818060 cites W2963341956 @default.
• W3103818060 cites W2963403868 @default.
• W3103818060 cites W2963506925 @default.
• W3103818060 cites W2963881719 @default.
• W3103818060 cites W2963919854 @default.
• W3103818060 cites W2964082031 @default.
• W3103818060 cites W2964121744 @default.
• W3103818060 cites W2964258094 @default.
• W3103818060 cites W2964308564 @default.
• W3103818060 cites W2965373594 @default.
• W3103818060 cites W2970294904 @default.
• W3103818060 cites W2970429618 @default.
• W3103818060 cites W2970521905 @default.
• W3103818060 cites W2970597249 @default.
• W3103818060 cites W2970868759 @default.
• W3103818060 cites W2995746049 @default.
• W3103818060 cites W2996285556 @default.
• W3103818060 cites W3035010485 @default.
• W3103818060 cites W3035160371 @default.
• W3103818060 cites W3037162118 @default.
• W3103818060 cites W3096648221 @default.
• W3103818060 doi "https://doi.org/10.18653/v1/2020.findings-emnlp.26" @default.
• W3103818060 hasPublicationYear "2020" @default.
• W3103818060 type Work @default.
• W3103818060 sameAs 3103818060 @default.
• W3103818060 citedByCount "6" @default.
• W3103818060 countsByYear W31038180602021 @default.
• W3103818060 countsByYear W31038180602022 @default.
• W3103818060 countsByYear W31038180602023 @default.
• W3103818060 crossrefType "proceedings-article" @default.
• W3103818060 hasAuthorship W3103818060A5002182453 @default.
• W3103818060 hasAuthorship W3103818060A5003760956 @default.
• W3103818060 hasAuthorship W3103818060A5005531754 @default.
• W3103818060 hasAuthorship W3103818060A5060910700 @default.
• W3103818060 hasAuthorship W3103818060A5066002127 @default.
• W3103818060 hasBestOaLocation W31038180601 @default.
• W3103818060 hasConcept C100660578 @default.
• W3103818060 hasConcept C111335779 @default.
• W3103818060 hasConcept C111368507 @default.
• W3103818060 hasConcept C111472728 @default.
• W3103818060 hasConcept C115961682 @default.
• W3103818060 hasConcept C119857082 @default.
• W3103818060 hasConcept C120665830 @default.
• W3103818060 hasConcept C121332964 @default.
• W3103818060 hasConcept C12725497 @default.
• W3103818060 hasConcept C127313418 @default.
• W3103818060 hasConcept C138885662 @default.
• W3103818060 hasConcept C145420912 @default.
• W3103818060 hasConcept C153180895 @default.
• W3103818060 hasConcept C154945302 @default.
• W3103818060 hasConcept C162324750 @default.
• W3103818060 hasConcept C163294075 @default.
• W3103818060 hasConcept C187736073 @default.
• W3103818060 hasConcept C192209626 @default.
• W3103818060 hasConcept C204321447 @default.
• W3103818060 hasConcept C2524010 @default.
• W3103818060 hasConcept C2776436953 @default.
• W3103818060 hasConcept C2777413886 @default.

• next