L'exacerbació dels estereotips (gramaticals) de gènere en la traducció automàtica anglès-castellà
Resum
En la traducció automàtica d'oracions aïllades a llengües que marquen el gènere, la informació necessària per seleccionar el gènere gramatical sovint n’és absent o és difícil d'extraure’n. Aquest estudi de caixa negra i centrat en el text mostra com la distribució de gènere del conjunt d’entrenament es distorsiona en l’eixida. Una avaluació humana revela que sovint no hi ha cap pista de gènere en l’original, cosa que condueix a traduccions estereotipades.
Paraules clau
traducció automàtica, castellà, anglès, gènere gramatical, distorsió de la distribució de gènereReferències
Basta, Christine; Costa-jussà, Marta R.; Fonollosa, José A. R. (2020). Towards mitigating gender bias in a decoder-based neural machine translation model by adding contextual information. In: Proceedings of the Fourth Widening Natural Language Processing Workshop, Seattle, U.S.A., July 2020, pp. 99–102. <https://aclanthology.org/2020.winlp-1.25>. [Accessed: 20221209].
Belinkov, Yonatan; Durrani, Nadir; Dalvi, Fahim; Sajjad, Hassan; Glass, James (2020). On the linguistic representational power of neural machine translation models. Computational Linguistics, v. 46, n. 1, pp. 1–52. <https://direct.mit.edu/coli/issue/46/1>. [Accessed: 20221209].
Gonen, Hila; Webster, Kellie. (2020). Automatically identifying gender issues in machine translation using perturbations. In: Findings of the Association for Computational Linguistics: EMNLP 2020, (November 16–20), pp. 1991–1995. Cornell University. <https://arxiv.org/abs/2004.14065>. [Accessed: 20221209].
Prates, Marcelo O. R.; Avelar, Pedro H.; Lamb, Luís C. (2020). Assessing gender bias in machine translation: a case study with Google Translate. Neural Computing and Applications, n. 32, pp. 6363–6381. <https://doi.org/10.1007/s00521-019-04144-6>. [Accessed: 20221209].
Rescigno, Argentina A.; Vanmassenhove, Eva; Monti, Johanna; Way, Andy (2021). A Case Study of Natural Gender Phenomena in Translation: A Comparison of Google Translate, Bing Microsoft Translator and DeepL for English to Italian, French and Spanish. In: Proceedings of the 7th Italian Conference on Computational Linguistics, CLiC-it 2020, in CEUR Workshop Proceedings vol. 2769, pp. 62-90. <https://aclanthology.org/2020.amta-impact.4> [Accessed: 20221209].
Renduchintala, Adithya; Diaz, Denise; Heafield, Kenneth; Li, Xian; Diab, Mona (2021). Gender Bias Amplification During Speed-Quality Optimization in Neural Machine Translation. In: Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers). Association for Computational Linguistics, pp. 99–109. <https://aclanthology.org/2021.acl-short.15>. [Accessed: 20221209].
Saunders, Danielle; Byrne, Bill (2020). Reducing gender bias in neural machine translation as a domain adaptation problem. In: Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics (ACL 2020, July 5–10). Association for Computational Linguistics, pp. 7724-7736. <https://aclanthology.org/2020.acl-main.690/>. [Accessed: 20221209].
Savoldi, Beatrice; Gaido, Marco; Bentivogli, Luisa; Negri, Mateo; Turchi, Marco (2021). Gender bias in machine translation. Transactions of the Association for Computational Linguistics, v. 9, pp. 845-874. <https://doi.org.10.1162/tacl_a_00401>. [Accessed: 20221209].
Štafanovičs, Arturs; Bergmanis, Toms; Pinnis, Mārcis (2020). Mitigating Gender Bias in Machine Translation with Target Gender Annotations. In: Proceedings of the Fifth Conference on Machine Translation. Association for Computational Linguistics, pp. 629–638. <https://aclanthology.org/2020.wmt-1.73>. [Accessed: 20221209].
Stanovsky, Gabriel; Smith, Noah A.; Zettlemoyer, Luke (2019). Evaluating gender bias in machine translation. In: Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics (ACL 2019, Florence, Italy, July 28–August 2, 2019), pp. 1679–1684. <https://doi.org/10.18653/v1/P19-1164>. [Accessed: 20221209].
Tomalin, Marcus; Byrne. Bill; Concannon, Shauna; Saunders, Danielle; Ullman, Stefanie (2021). The practical ethics of bias reduction in machine translation: why domain adaptation is better than data debiasing. Ethics and Information Technology, n. 23, pp. 419-433. <https://doi.org/10.1007/s10676-021-09583-1>. [Accessed: 20221209].
Toral, Antonio. (2019). Post-editese: an Exacerbated Translationese. In: Proceedings of Machine Translation Summit XVII: Research Track, Dublin, Ireland. European Association for Machine Translation, pp. 273–281. <https://aclanthology.org/W19-6627/>. [Accessed: 20221209].
Vanmassenhove, Eva; Hardmeier, Christian; Way, Andy (2018). Getting Gender Right in Neural Machine Translation. In: Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, Brussels, Belgium. Association for Computational Linguistics, pp. 3003–3008. <https://aclanthology.org/D18-1334>. [Accessed: 20221209].
Vanmassenhove, Eva; Shterionov, Dimitar; Way, Andy (2019). Lost in Translation: Loss and Decay of Linguistic Richness in Machine Translation. In: Proceedings of Machine Translation Summit XVII: Research Track, Dublin, Ireland. European Association for Machine Translation, pp. 222.232. <https://aclanthology.org/W19-6622>. [Accessed: 20221209].
Vaswani, Ashish; Shazeer, Noam; Parmar, Niki; Uszkoreit, Jakob; Jones, Llion; Gomez, Aidan N.; Kaiser, Łukasz; Polosukhin, Illia (2019). Attention is all you need. In: Conference on Neural Information Processing Systems (NIPS 2017), pp. 1-11. <https://proceedings.neurips.cc/paper/2017/file/3f5ee243547dee91fbd053c1c4a845aa-Paper.pdf>. [Accessed: 20221209].
Volansky, Vered; Ordan, Noam; Wintner, Shuly (2015). On the features of translationese. Digital Scholarship in the Humanities, v. 30, n. 1, pp. 98-118. <https://doi.org/10.1093/llc/fqt031>. [Accessed: 20221209].
Publicades
Descàrregues
Drets d'autor (c) 2022 Nerea Ondoño-Soler, Mikel L. Forcada
Aquesta obra està sota una llicència internacional Creative Commons Reconeixement 4.0.
