FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3207182695> ?p ?o ?g. }

• W3207182695 abstract "Targeted evaluations have found that machine translation systems often output incorrect gender in translations, even when the gender is clear from context. Furthermore, these incorrectly gendered translations have the potential to reflect or amplify social biases. We propose gender-filtered self-training (GFST) to improve gender translation accuracy on unambiguously gendered inputs. Our GFST approach uses a source monolingual corpus and an initial model to generate gender-specific pseudo-parallel corpora which are then filtered and added to the training data. We evaluate GFST on translation from English into five languages, finding that it improves gender accuracy without damaging generic quality. We also show the viability of GFST on several experimental settings, including re-training from scratch, fine-tuning, controlling the gender balance of the data, forward translation, and back-translation." @default.
• W3207182695 created "2021-10-25" @default.
• W3207182695 creator A5057049978 @default.
• W3207182695 creator A5057417152 @default.
• W3207182695 creator A5090357305 @default.
• W3207182695 creator A5091384195 @default.
• W3207182695 date "2021-01-01" @default.
• W3207182695 modified "2023-09-25" @default.
• W3207182695 title "GFST: Gender-Filtered Self-Training for More Accurate Gender in Translation" @default.
• W3207182695 cites W1601432161 @default.
• W3207182695 cites W2095705004 @default.
• W3207182695 cites W2101456909 @default.
• W3207182695 cites W2148708890 @default.
• W3207182695 cites W22168010 @default.
• W3207182695 cites W2483215953 @default.
• W3207182695 cites W2512924740 @default.
• W3207182695 cites W2561274697 @default.
• W3207182695 cites W2567571499 @default.
• W3207182695 cites W2744813330 @default.
• W3207182695 cites W2792115266 @default.
• W3207182695 cites W2889326796 @default.
• W3207182695 cites W2891525068 @default.
• W3207182695 cites W2903193068 @default.
• W3207182695 cites W2933138175 @default.
• W3207182695 cites W2945700568 @default.
• W3207182695 cites W2950888501 @default.
• W3207182695 cites W2952328691 @default.
• W3207182695 cites W2962784628 @default.
• W3207182695 cites W2963216553 @default.
• W3207182695 cites W2963403868 @default.
• W3207182695 cites W2963457723 @default.
• W3207182695 cites W2963526187 @default.
• W3207182695 cites W2963532001 @default.
• W3207182695 cites W2963807318 @default.
• W3207182695 cites W2964121744 @default.
• W3207182695 cites W2970311224 @default.
• W3207182695 cites W2972568911 @default.
• W3207182695 cites W2972680990 @default.
• W3207182695 cites W2972866455 @default.
• W3207182695 cites W2972972637 @default.
• W3207182695 cites W3034458735 @default.
• W3207182695 cites W3035070478 @default.
• W3207182695 cites W3037697022 @default.
• W3207182695 cites W3037831233 @default.
• W3207182695 cites W3041866211 @default.
• W3207182695 cites W3082714809 @default.
• W3207182695 cites W3093818688 @default.
• W3207182695 cites W3101004475 @default.
• W3207182695 cites W3119159844 @default.
• W3207182695 cites W3119161189 @default.
• W3207182695 cites W3133764719 @default.
• W3207182695 cites W3195725782 @default.
• W3207182695 cites W3202218022 @default.
• W3207182695 doi "https://doi.org/10.18653/v1/2021.emnlp-main.123" @default.
• W3207182695 hasPublicationYear "2021" @default.
• W3207182695 type Work @default.
• W3207182695 sameAs 3207182695 @default.
• W3207182695 citedByCount "0" @default.
• W3207182695 crossrefType "proceedings-article" @default.
• W3207182695 hasAuthorship W3207182695A5057049978 @default.
• W3207182695 hasAuthorship W3207182695A5057417152 @default.
• W3207182695 hasAuthorship W3207182695A5090357305 @default.
• W3207182695 hasAuthorship W3207182695A5091384195 @default.
• W3207182695 hasBestOaLocation W32071826951 @default.
• W3207182695 hasConcept C104317684 @default.
• W3207182695 hasConcept C105580179 @default.
• W3207182695 hasConcept C107993555 @default.
• W3207182695 hasConcept C111472728 @default.
• W3207182695 hasConcept C111919701 @default.
• W3207182695 hasConcept C121332964 @default.
• W3207182695 hasConcept C138885662 @default.
• W3207182695 hasConcept C144024400 @default.
• W3207182695 hasConcept C149364088 @default.
• W3207182695 hasConcept C151730666 @default.
• W3207182695 hasConcept C153294291 @default.
• W3207182695 hasConcept C154945302 @default.
• W3207182695 hasConcept C15744967 @default.
• W3207182695 hasConcept C168031717 @default.
• W3207182695 hasConcept C169760540 @default.
• W3207182695 hasConcept C185592680 @default.
• W3207182695 hasConcept C203005215 @default.
• W3207182695 hasConcept C204321447 @default.
• W3207182695 hasConcept C2777211547 @default.
• W3207182695 hasConcept C2779343474 @default.
• W3207182695 hasConcept C2779530757 @default.
• W3207182695 hasConcept C2781235140 @default.
• W3207182695 hasConcept C2993152109 @default.
• W3207182695 hasConcept C41008148 @default.
• W3207182695 hasConcept C51632099 @default.
• W3207182695 hasConcept C55493867 @default.
• W3207182695 hasConcept C86803240 @default.
• W3207182695 hasConceptScore W3207182695C104317684 @default.
• W3207182695 hasConceptScore W3207182695C105580179 @default.
• W3207182695 hasConceptScore W3207182695C107993555 @default.
• W3207182695 hasConceptScore W3207182695C111472728 @default.
• W3207182695 hasConceptScore W3207182695C111919701 @default.
• W3207182695 hasConceptScore W3207182695C121332964 @default.
• W3207182695 hasConceptScore W3207182695C138885662 @default.
• W3207182695 hasConceptScore W3207182695C144024400 @default.
• W3207182695 hasConceptScore W3207182695C149364088 @default.

• next