Matches in SemOpenAlex for { <https://semopenalex.org/work/W4320497798> ?p ?o ?g. }
Showing items 1 to 95 of
95
with 100 items per page.
- W4320497798 endingPage "26" @default.
- W4320497798 startingPage "1" @default.
- W4320497798 abstract "It is hard to evaluate translations objectively and accurately, which limits the applications of machine translation. In this article, we assume that the above phenomenon is caused by noise interference during translation evaluation, and we handle the problem through a perspective of causal inference. We assume that the observable translation score is affected by the unobservable true translation quality and some noise simultaneously. If there is a variable that is related to the noise and independent to the true translation quality, the related noise can be eliminated by removing the effect of that variable from the observed score. Based on the above causality hypothesis, this article studies the length bias problem of beam search for neural machine translation (NMT) and the input related noise problem of translation quality estimation (QE). For the NMT length bias problem, we conduct the experiments on four typical NMT tasks (Uyghur–Chinese, Chinese–English, English–German, and English–French) with different scales of datasets. Comparing with previous approaches, the proposed causal motivated method is model-agnostic and does not require supervised training. For QE tasks, we conduct the experiments on the WMT’20 submissions. Experimental results show that the denoised QE results gain better Pearson’s correlation scores with human assessed scores compared to the original submissions. Further analyses on the NMT and QE tasks also demonstrate the rationality of the empirical assumptions made on our methods." @default.
- W4320497798 created "2023-02-14" @default.
- W4320497798 creator A5058216405 @default.
- W4320497798 creator A5073550269 @default.
- W4320497798 creator A5075237838 @default.
- W4320497798 creator A5087631670 @default.
- W4320497798 date "2023-05-09" @default.
- W4320497798 modified "2023-10-17" @default.
- W4320497798 title "Approximating to the Real Translation Quality for Neural Machine Translation via Causal Motivated Methods" @default.
- W4320497798 cites W2087039607 @default.
- W4320497798 cites W2124807415 @default.
- W4320497798 cites W2144746247 @default.
- W4320497798 cites W2146574666 @default.
- W4320497798 cites W2154124206 @default.
- W4320497798 cites W2252272516 @default.
- W4320497798 cites W2470207857 @default.
- W4320497798 cites W2757222607 @default.
- W4320497798 cites W2933138175 @default.
- W4320497798 cites W2962784628 @default.
- W4320497798 cites W2963212250 @default.
- W4320497798 cites W2963382396 @default.
- W4320497798 cites W2963506925 @default.
- W4320497798 cites W2963665552 @default.
- W4320497798 cites W2963979492 @default.
- W4320497798 cites W2964199361 @default.
- W4320497798 cites W2990461750 @default.
- W4320497798 cites W2997591727 @default.
- W4320497798 cites W3035390927 @default.
- W4320497798 doi "https://doi.org/10.1145/3583684" @default.
- W4320497798 hasPublicationYear "2023" @default.
- W4320497798 type Work @default.
- W4320497798 citedByCount "0" @default.
- W4320497798 crossrefType "journal-article" @default.
- W4320497798 hasAuthorship W4320497798A5058216405 @default.
- W4320497798 hasAuthorship W4320497798A5073550269 @default.
- W4320497798 hasAuthorship W4320497798A5075237838 @default.
- W4320497798 hasAuthorship W4320497798A5087631670 @default.
- W4320497798 hasBestOaLocation W43204977981 @default.
- W4320497798 hasConcept C104317684 @default.
- W4320497798 hasConcept C105580179 @default.
- W4320497798 hasConcept C111472728 @default.
- W4320497798 hasConcept C115961682 @default.
- W4320497798 hasConcept C119857082 @default.
- W4320497798 hasConcept C138885662 @default.
- W4320497798 hasConcept C149364088 @default.
- W4320497798 hasConcept C149782125 @default.
- W4320497798 hasConcept C154945302 @default.
- W4320497798 hasConcept C185592680 @default.
- W4320497798 hasConcept C203005215 @default.
- W4320497798 hasConcept C2776214188 @default.
- W4320497798 hasConcept C2779530757 @default.
- W4320497798 hasConcept C2780695315 @default.
- W4320497798 hasConcept C33923547 @default.
- W4320497798 hasConcept C41008148 @default.
- W4320497798 hasConcept C55493867 @default.
- W4320497798 hasConcept C99498987 @default.
- W4320497798 hasConceptScore W4320497798C104317684 @default.
- W4320497798 hasConceptScore W4320497798C105580179 @default.
- W4320497798 hasConceptScore W4320497798C111472728 @default.
- W4320497798 hasConceptScore W4320497798C115961682 @default.
- W4320497798 hasConceptScore W4320497798C119857082 @default.
- W4320497798 hasConceptScore W4320497798C138885662 @default.
- W4320497798 hasConceptScore W4320497798C149364088 @default.
- W4320497798 hasConceptScore W4320497798C149782125 @default.
- W4320497798 hasConceptScore W4320497798C154945302 @default.
- W4320497798 hasConceptScore W4320497798C185592680 @default.
- W4320497798 hasConceptScore W4320497798C203005215 @default.
- W4320497798 hasConceptScore W4320497798C2776214188 @default.
- W4320497798 hasConceptScore W4320497798C2779530757 @default.
- W4320497798 hasConceptScore W4320497798C2780695315 @default.
- W4320497798 hasConceptScore W4320497798C33923547 @default.
- W4320497798 hasConceptScore W4320497798C41008148 @default.
- W4320497798 hasConceptScore W4320497798C55493867 @default.
- W4320497798 hasConceptScore W4320497798C99498987 @default.
- W4320497798 hasFunder F4320321001 @default.
- W4320497798 hasIssue "5" @default.
- W4320497798 hasLocation W43204977981 @default.
- W4320497798 hasOpenAccess W4320497798 @default.
- W4320497798 hasPrimaryLocation W43204977981 @default.
- W4320497798 hasRelatedWork W2103288873 @default.
- W4320497798 hasRelatedWork W2313858059 @default.
- W4320497798 hasRelatedWork W2509777158 @default.
- W4320497798 hasRelatedWork W2807995432 @default.
- W4320497798 hasRelatedWork W2963058055 @default.
- W4320497798 hasRelatedWork W3022579377 @default.
- W4320497798 hasRelatedWork W3192376505 @default.
- W4320497798 hasRelatedWork W4298137069 @default.
- W4320497798 hasRelatedWork W596611847 @default.
- W4320497798 hasRelatedWork W97426361 @default.
- W4320497798 hasVolume "22" @default.
- W4320497798 isParatext "false" @default.
- W4320497798 isRetracted "false" @default.
- W4320497798 workType "article" @default.