Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891520362> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2891520362 abstract "In this work we introduce dual conditional cross-entropy filtering for noisy parallel data. For each sentence pair of the noisy parallel corpus we compute cross-entropy scores according to two inverse translation models trained on clean data. We penalize divergent cross-entropies and weigh the penalty by the cross-entropy average of both models. Sorting or thresholding according to these scores results in better subsets of parallel data. We achieve higher BLEU scores with models trained on parallel data filtered only from Paracrawl than with models trained on clean WMT data. We further evaluate our method in the context of the WMT2018 shared task on parallel corpus filtering and achieve the overall highest ranking scores of the shared task, scoring top in three out of four subtasks." @default.
- W2891520362 created "2018-09-27" @default.
- W2891520362 creator A5065193578 @default.
- W2891520362 date "2018-09-01" @default.
- W2891520362 modified "2023-09-27" @default.
- W2891520362 title "Dual Conditional Cross-Entropy Filtering of Noisy Parallel Corpora" @default.
- W2891520362 cites W1905522558 @default.
- W2891520362 cites W2117278770 @default.
- W2891520362 cites W2767899794 @default.
- W2891520362 cites W2794365787 @default.
- W2891520362 cites W2798389157 @default.
- W2891520362 cites W2902918014 @default.
- W2891520362 cites W2950359962 @default.
- W2891520362 cites W2963266340 @default.
- W2891520362 cites W2963403868 @default.
- W2891520362 cites W2963919854 @default.
- W2891520362 hasPublicationYear "2018" @default.
- W2891520362 type Work @default.
- W2891520362 sameAs 2891520362 @default.
- W2891520362 citedByCount "2" @default.
- W2891520362 countsByYear W28915203622019 @default.
- W2891520362 countsByYear W28915203622020 @default.
- W2891520362 crossrefType "posted-content" @default.
- W2891520362 hasAuthorship W2891520362A5065193578 @default.
- W2891520362 hasConcept C101721835 @default.
- W2891520362 hasConcept C106301342 @default.
- W2891520362 hasConcept C115961682 @default.
- W2891520362 hasConcept C119857082 @default.
- W2891520362 hasConcept C121332964 @default.
- W2891520362 hasConcept C153180895 @default.
- W2891520362 hasConcept C154945302 @default.
- W2891520362 hasConcept C167981619 @default.
- W2891520362 hasConcept C189430467 @default.
- W2891520362 hasConcept C191178318 @default.
- W2891520362 hasConcept C203005215 @default.
- W2891520362 hasConcept C204321447 @default.
- W2891520362 hasConcept C2777530160 @default.
- W2891520362 hasConcept C41008148 @default.
- W2891520362 hasConcept C62520636 @default.
- W2891520362 hasConcept C9679016 @default.
- W2891520362 hasConceptScore W2891520362C101721835 @default.
- W2891520362 hasConceptScore W2891520362C106301342 @default.
- W2891520362 hasConceptScore W2891520362C115961682 @default.
- W2891520362 hasConceptScore W2891520362C119857082 @default.
- W2891520362 hasConceptScore W2891520362C121332964 @default.
- W2891520362 hasConceptScore W2891520362C153180895 @default.
- W2891520362 hasConceptScore W2891520362C154945302 @default.
- W2891520362 hasConceptScore W2891520362C167981619 @default.
- W2891520362 hasConceptScore W2891520362C189430467 @default.
- W2891520362 hasConceptScore W2891520362C191178318 @default.
- W2891520362 hasConceptScore W2891520362C203005215 @default.
- W2891520362 hasConceptScore W2891520362C204321447 @default.
- W2891520362 hasConceptScore W2891520362C2777530160 @default.
- W2891520362 hasConceptScore W2891520362C41008148 @default.
- W2891520362 hasConceptScore W2891520362C62520636 @default.
- W2891520362 hasConceptScore W2891520362C9679016 @default.
- W2891520362 hasLocation W28915203621 @default.
- W2891520362 hasOpenAccess W2891520362 @default.
- W2891520362 hasPrimaryLocation W28915203621 @default.
- W2891520362 hasRelatedWork W1519512946 @default.
- W2891520362 hasRelatedWork W1887786044 @default.
- W2891520362 hasRelatedWork W1986468312 @default.
- W2891520362 hasRelatedWork W2086846715 @default.
- W2891520362 hasRelatedWork W2148922863 @default.
- W2891520362 hasRelatedWork W2162583378 @default.
- W2891520362 hasRelatedWork W2177717508 @default.
- W2891520362 hasRelatedWork W2250890374 @default.
- W2891520362 hasRelatedWork W2433238124 @default.
- W2891520362 hasRelatedWork W2550969009 @default.
- W2891520362 hasRelatedWork W2560567690 @default.
- W2891520362 hasRelatedWork W2741442884 @default.
- W2891520362 hasRelatedWork W2945405384 @default.
- W2891520362 hasRelatedWork W2951985405 @default.
- W2891520362 hasRelatedWork W2963281280 @default.
- W2891520362 hasRelatedWork W2970155092 @default.
- W2891520362 hasRelatedWork W3028222090 @default.
- W2891520362 hasRelatedWork W3129146692 @default.
- W2891520362 hasRelatedWork W2135919823 @default.
- W2891520362 hasRelatedWork W3131802060 @default.
- W2891520362 isParatext "false" @default.
- W2891520362 isRetracted "false" @default.
- W2891520362 magId "2891520362" @default.
- W2891520362 workType "article" @default.