Matches in SemOpenAlex for { <https://semopenalex.org/work/W3194060937> ?p ?o ?g. }
Showing items 1 to 74 of
74
with 100 items per page.
- W3194060937 abstract "Most state-of-the-art QE systems built upon neural networks have achieved promising performances on benchmark datasets. However, the performance of these methods can be easily influenced by the inherent features of the model input, such as the length of input sequence or the number of unseen tokens. In this paper, we introduce a causal inference based method to eliminate the negative impact caused by the characters of the input for a QE system. Specifically, we propose an iterative denoising framework for multiple confounding features. The confounder elimination operation at each iteration step is implemented by a Half-Sibling Regression based method. We conduct our experiments on the official datasets and submissions from WMT 2020 Quality Estimation Shared Task of Sentence-Level Direct Assessment. Experimental results show that the denoised QE results gain better Pearson’s correlation scores with human assessments compared to the original submissions." @default.
- W3194060937 created "2021-08-30" @default.
- W3194060937 creator A5058216405 @default.
- W3194060937 creator A5073550269 @default.
- W3194060937 creator A5075237838 @default.
- W3194060937 creator A5087631670 @default.
- W3194060937 date "2021-01-01" @default.
- W3194060937 modified "2023-10-17" @default.
- W3194060937 title "Removing Input Confounder for Translation Quality Estimation via a Causal Motivated Method" @default.
- W3194060937 cites W2470207857 @default.
- W3194060937 cites W2963506925 @default.
- W3194060937 doi "https://doi.org/10.1007/978-3-030-85896-4_28" @default.
- W3194060937 hasPublicationYear "2021" @default.
- W3194060937 type Work @default.
- W3194060937 sameAs 3194060937 @default.
- W3194060937 citedByCount "0" @default.
- W3194060937 crossrefType "book-chapter" @default.
- W3194060937 hasAuthorship W3194060937A5058216405 @default.
- W3194060937 hasAuthorship W3194060937A5073550269 @default.
- W3194060937 hasAuthorship W3194060937A5075237838 @default.
- W3194060937 hasAuthorship W3194060937A5087631670 @default.
- W3194060937 hasConcept C105795698 @default.
- W3194060937 hasConcept C119857082 @default.
- W3194060937 hasConcept C124101348 @default.
- W3194060937 hasConcept C13280743 @default.
- W3194060937 hasConcept C153180895 @default.
- W3194060937 hasConcept C154945302 @default.
- W3194060937 hasConcept C158600405 @default.
- W3194060937 hasConcept C162324750 @default.
- W3194060937 hasConcept C185798385 @default.
- W3194060937 hasConcept C187736073 @default.
- W3194060937 hasConcept C203005215 @default.
- W3194060937 hasConcept C205649164 @default.
- W3194060937 hasConcept C2776214188 @default.
- W3194060937 hasConcept C2777530160 @default.
- W3194060937 hasConcept C2780451532 @default.
- W3194060937 hasConcept C33923547 @default.
- W3194060937 hasConcept C41008148 @default.
- W3194060937 hasConcept C83546350 @default.
- W3194060937 hasConceptScore W3194060937C105795698 @default.
- W3194060937 hasConceptScore W3194060937C119857082 @default.
- W3194060937 hasConceptScore W3194060937C124101348 @default.
- W3194060937 hasConceptScore W3194060937C13280743 @default.
- W3194060937 hasConceptScore W3194060937C153180895 @default.
- W3194060937 hasConceptScore W3194060937C154945302 @default.
- W3194060937 hasConceptScore W3194060937C158600405 @default.
- W3194060937 hasConceptScore W3194060937C162324750 @default.
- W3194060937 hasConceptScore W3194060937C185798385 @default.
- W3194060937 hasConceptScore W3194060937C187736073 @default.
- W3194060937 hasConceptScore W3194060937C203005215 @default.
- W3194060937 hasConceptScore W3194060937C205649164 @default.
- W3194060937 hasConceptScore W3194060937C2776214188 @default.
- W3194060937 hasConceptScore W3194060937C2777530160 @default.
- W3194060937 hasConceptScore W3194060937C2780451532 @default.
- W3194060937 hasConceptScore W3194060937C33923547 @default.
- W3194060937 hasConceptScore W3194060937C41008148 @default.
- W3194060937 hasConceptScore W3194060937C83546350 @default.
- W3194060937 hasLocation W31940609371 @default.
- W3194060937 hasOpenAccess W3194060937 @default.
- W3194060937 hasPrimaryLocation W31940609371 @default.
- W3194060937 hasRelatedWork W11849241 @default.
- W3194060937 hasRelatedWork W11991885 @default.
- W3194060937 hasRelatedWork W1224884 @default.
- W3194060937 hasRelatedWork W12829028 @default.
- W3194060937 hasRelatedWork W14516383 @default.
- W3194060937 hasRelatedWork W6266756 @default.
- W3194060937 hasRelatedWork W728297 @default.
- W3194060937 hasRelatedWork W7724241 @default.
- W3194060937 hasRelatedWork W8197146 @default.
- W3194060937 hasRelatedWork W9761094 @default.
- W3194060937 isParatext "false" @default.
- W3194060937 isRetracted "false" @default.
- W3194060937 magId "3194060937" @default.
- W3194060937 workType "book-chapter" @default.