9/11/2023 0 Comments Ender language translatorFurthermore, these incorrectly gendered translations have the potential to reflect or amplify social biases. Targeted evaluations have found that machine translation systems often output incorrect gender in translations, even when the gender is clear from context. Online and Punta Cana, Dominican Republic Proceedings of the 2021 Conference on Empirical Methods in Natural Language ProcessingĪssociation for Computational Linguistics GFST: Gender-Filtered Self-Training for More Accurate Gender in Translation 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.", ![]() We evaluate GFST on translation from English into five languages, finding that it improves gender accuracy without damaging generic quality. 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 propose gender-filtered self-training (GFST) to improve gender translation accuracy on unambiguously gendered inputs. Publisher = "Association for Computational Linguistics",Ībstract = "Targeted evaluations have found that machine translation systems often output incorrect gender in translations, even when the gender is clear from context. Cite (Informal): GFST: Gender-Filtered Self-Training for More Accurate Gender in Translation (Choubey et al., EMNLP 2021) Copy Citation: BibTeX Markdown MODS XML Endnote More options… PDF: Video: Code = "ender-Filtered Self-Training for More Accurate Gender in Translation",īooktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",Īddress = "Online and Punta Cana, Dominican Republic", Association for Computational Linguistics. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 1640–1654, Online and Punta Cana, Dominican Republic. ![]() GFST: Gender-Filtered Self-Training for More Accurate Gender in Translation. Anthology ID: 2021.emnlp-main.123 Volume: Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing Month: November Year: 2021 Address: Online and Punta Cana, Dominican Republic Venue: EMNLP SIG: Publisher: Association for Computational Linguistics Note: Pages: 1640–1654 Language: URL: DOI: 10.18653/v1/2021.emnlp-main.123 Bibkey: choubey-etal-2021-gfst Cite (ACL): Prafulla Kumar Choubey, Anna Currey, Prashant Mathur, and Georgiana Dinu. 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. Abstract Targeted evaluations have found that machine translation systems often output incorrect gender in translations, even when the gender is clear from context.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |