Papers
Topics
Authors
Recent
Search
2000 character limit reached

Optimizing example selection for retrieval-augmented machine translation with translation memories

Published 23 May 2024 in cs.CL | (2405.15070v1)

Abstract: Retrieval-augmented machine translation leverages examples from a translation memory by retrieving similar instances. These examples are used to condition the predictions of a neural decoder. We aim to improve the upstream retrieval step and consider a fixed downstream edit-based model: the multi-Levenshtein Transformer. The task consists of finding a set of examples that maximizes the overall coverage of the source sentence. To this end, we rely on the theory of submodular functions and explore new algorithms to optimize this coverage. We evaluate the resulting performance gains for the machine translation task.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (47)
  1. In-context examples selection for machine translation. In A. Rogers, J. Boyd-Graber & N. Okazaki, Éds., Findings of the Association for Computational Linguistics: ACL 2023, p. 8857–8873, Toronto, Canada: Association for Computational Linguistics. doi : 10.18653/v1/2023.findings-acl.564.
  2. Bawden R. & Yvon F. (2023). Investigating the translation performance of a large multilingual language model: the case of BLOOM. In M. Nurminen, J. Brenner, M. Koponen, S. Latomaa, M. Mikhailov, F. Schierl, T. Ranasinghe, E. Vanmassenhove, S. A. Vidal, N. Aranberri, M. Nunziatini, C. P. Escartín, M. Forcada, M. Popovic, C. Scarton & H. Moniz, Éds., Proceedings of the 24th Annual Conference of the European Association for Machine Translation, p. 157–170, Tampere, Finland: European Association for Machine Translation.
  3. Bouthors M., Crego J. & Yvon F. (2023). Towards example-based NMT with multi-Levenshtein transformers. In H. Bouamor, J. Pino & K. Bali, Éds., Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, p. 1830–1846, Singapore: Association for Computational Linguistics. doi : 10.18653/v1/2023.emnlp-main.113.
  4. Bowker L. (2002). Computer-aided translation technology: A practical introduction. University of Ottawa Press.
  5. Bulte B. & Tezcan A. (2019). Neural fuzzy repair: Integrating fuzzy matches into neural machine translation. In A. Korhonen, D. Traum & L. Màrquez, Éds., Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, p. 1800–1809, Florence, Italy: Association for Computational Linguistics. doi : 10.18653/v1/P19-1175.
  6. Neural machine translation with monolingual translation memory. In C. Zong, F. Xia, W. Li & R. Navigli, Éds., Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), p. 7307–7318, Online: Association for Computational Linguistics. doi : 10.18653/v1/2021.acl-long.567.
  7. Carl M., Way A. & Daelemans W. (2004). Recent advances in example-based machine translation. Computational Linguistics, 30, 516–520. doi : 10.1162/0891201042544866.
  8. Neural machine translation with contrastive translation memories. In Y. Goldberg, Z. Kozareva & Y. Zhang, Éds., Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, p. 3591–3601, Abu Dhabi, United Arab Emirates: Association for Computational Linguistics. doi : 10.18653/v1/2022.emnlp-main.235.
  9. Daumé H., Langford J. & Marcu D. (2009). Search-based structured prediction. Machine Learning, 75(3), 297–325. doi : 10.1007/s10994-009-5106-x.
  10. Multi-domain neural machine translation through unsupervised adaptation. In O. Bojar, C. Buck, R. Chatterjee, C. Federmann, Y. Graham, B. Haddow, M. Huck, A. J. Yepes, P. Koehn & J. Kreutzer, Éds., Proceedings of the Second Conference on Machine Translation, p. 127–137, Copenhagen, Denmark: Association for Computational Linguistics. doi : 10.18653/v1/W17-4713.
  11. Goldstein J. & Carbonell J. (1998). Summarization: (1) using MMR for diversity- based reranking and (2) evaluating summaries. In TIPSTER TEXT PROGRAM PHASE III: Proceedings of a Workshop held at Baltimore, Maryland, October 13-15, 1998, p. 181–195, Baltimore, Maryland, USA: Association for Computational Linguistics. doi : 10.3115/1119089.1119120.
  12. Gu J., Wang C. & Zhao J. (2019). Levenshtein transformer. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d'Alché-Buc, E. Fox & R. Garnett, Éds., Advances in Neural Information Processing Systems, volume 32: Curran Associates, Inc.
  13. Search Engine Guided Neural Machine Translation. Proceedings of the AAAI Conference on Artificial Intelligence, 32(1). doi : 10.1609/aaai.v32i1.12013.
  14. Gupta S., Gardner M. & Singh S. (2023). Coverage-based example selection for in-context learning. In H. Bouamor, J. Pino & K. Bali, Éds., Findings of the Association for Computational Linguistics: EMNLP 2023, p. 13924–13950, Singapore: Association for Computational Linguistics. doi : 10.18653/v1/2023.findings-emnlp.930.
  15. He J., Neubig G. & Berg-Kirkpatrick T. (2021a). Efficient nearest neighbor language models. In M.-F. Moens, X. Huang, L. Specia & S. W.-t. Yih, Éds., Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, p. 5703–5714, Online and Punta Cana, Dominican Republic: Association for Computational Linguistics. doi : 10.18653/v1/2021.emnlp-main.461.
  16. Fast and accurate neural machine translation with translation memory. In C. Zong, F. Xia, W. Li & R. Navigli, Éds., Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), p. 3170–3180, Online: Association for Computational Linguistics. doi : 10.18653/v1/2021.acl-long.246.
  17. How good are GPT models at machine translation? a comprehensive evaluation. CoRR, abs/2302.09210. doi : 10.48550/ARXIV.2302.09210.
  18. Improving Retrieval Augmented Neural Machine Translation by Controlling Source and Fuzzy-Match Interactions. arXiv:2210.05047 [cs], doi : 10.48550/arXiv.2210.05047.
  19. Nearest Neighbor Machine Translation. In Proceedings of the International Conference on Learning Representations.
  20. Knyazeva E., Wisniewski G. & Yvon F. (2018). Les méthodes « apprendre à chercher » en traitement automatique des langues : un état de l’art [a survey of learning-to-search techniques in natural language processing]. Traitement Automatique des Langues, 59(1), 39–63.
  21. Koehn P. & Senellart J. (2010). Convergence of translation memory and statistical machine translation. In V. Zhechev, Éd., Proceedings of the Second Joint EM+/CNGL Workshop: Bringing MT to the User: Research on Integrating MT in the Translation Industry, p. 21–32, Denver, Colorado, USA: Association for Machine Translation in the Americas.
  22. Krause A. & Golovin D. (2014). Submodular function maximization. In L. Bordeaux, Y. Hamadi & P. Kohli, Éds., Tractability: Practical Approaches to Hard Problems, p. 71–104. Cambridge University Press.
  23. A survey on retrieval-augmented text generation. CoRR, abs/2202.01110.
  24. Lin H. & Bilmes J. (2011). A class of submodular functions for document summarization. In D. Lin, Y. Matsumoto & R. Mihalcea, Éds., Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics: Human Language Technologies, p. 510–520, Portland, Oregon, USA: Association for Computational Linguistics.
  25. CTQScorer: Combining multiple features for in-context example selection for machine translation. In H. Bouamor, J. Pino & K. Bali, Éds., Findings of the Association for Computational Linguistics: EMNLP 2023, p. 7736–7752, Singapore: Association for Computational Linguistics. doi : 10.18653/v1/2023.findings-emnlp.519.
  26. Martins P. H., Marinho Z. & Martins A. F. T. (2022). Chunk-based nearest neighbor machine translation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, p. 4228–4245, Abu Dhabi, United Arab Emirates: Association for Computational Linguistics. doi : https://aclanthology.org/emnlp-22/2022.emnlp-main.284.
  27. Fast nearest neighbor machine translation. In S. Muresan, P. Nakov & A. Villavicencio, Éds., Findings of the Association for Computational Linguistics: ACL 2022, p. 555–565, Dublin, Ireland: Association for Computational Linguistics. doi : 10.18653/v1/2022.findings-acl.47.
  28. Adaptive machine translation with large language models. In M. Nurminen, J. Brenner, M. Koponen, S. Latomaa, M. Mikhailov, F. Schierl, T. Ranasinghe, E. Vanmassenhove, S. A. Vidal, N. Aranberri, M. Nunziatini, C. P. Escartín, M. Forcada, M. Popovic, C. Scarton & H. Moniz, Éds., Proceedings of the 24th Annual Conference of the European Association for Machine Translation, p. 227–237, Tampere, Finland: European Association for Machine Translation.
  29. Augmenting large language model translators via translation memories. In A. Rogers, J. Boyd-Graber & N. Okazaki, Éds., Findings of the Association for Computational Linguistics: ACL 2023, p. 10287–10299, Toronto, Canada: Association for Computational Linguistics. doi : 10.18653/v1/2023.findings-acl.653.
  30. Nagao M. (1984). A framework of a mechanical translation between Japanese and English by analogy principle. In A. Elithorn & R. Banerji, Éds., Artificial and human intelligence: Elsevier Science Publishers. B.V.
  31. Niwa A., Takase S. & Okazaki N. (2022). Nearest neighbor non-autoregressive text generation. CoRR, abs/2208.12496. doi : 10.48550/ARXIV.2208.12496.
  32. Bleu: a method for automatic evaluation of machine translation. In P. Isabelle, E. Charniak & D. Lin, Éds., Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, p. 311–318, Philadelphia, Pennsylvania, USA: Association for Computational Linguistics. doi : 10.3115/1073083.1073135.
  33. Priming neural machine translation. In L. Barrault, O. Bojar, F. Bougares, R. Chatterjee, M. R. Costa-jussà, C. Federmann, M. Fishel, A. Fraser, Y. Graham, P. Guzman, B. Haddow, M. Huck, A. J. Yepes, P. Koehn, A. Martins, M. Morishita, C. Monz, M. Nagata, T. Nakazawa & M. Negri, Éds., Proceedings of the Fifth Conference on Machine Translation, p. 516–527, Online: Association for Computational Linguistics.
  34. Post M. (2018). A call for clarity in reporting BLEU scores. In O. Bojar, R. Chatterjee, C. Federmann, M. Fishel, Y. Graham, B. Haddow, M. Huck, A. J. Yepes, P. Koehn, C. Monz, M. Negri, A. Névéol, M. Neves, M. Post, L. Specia, M. Turchi & K. Verspoor, Éds., Proceedings of the Third Conference on Machine Translation: Research Papers, p. 186–191, Brussels, Belgium: Association for Computational Linguistics. doi : 10.18653/v1/W18-6319.
  35. Robertson S. E. & Jones K. S. (1976). Relevance weighting of search terms. Journal of the American Society for Information Science, 27(3), 129–146. doi : https://doi.org/10.1002/asi.4630270302.
  36. Ross S., Gordon G. & Bagnell D. (2011). A reduction of imitation learning and structured prediction to no-regret online learning. In G. Gordon, D. Dunson & M. Dudík, Éds., Proceedings of the Fourteenth International Conference on Artificial Intelligence and Statistics, volume 15 de Proceedings of Machine Learning Research, p. 627–635, Fort Lauderdale, FL, USA: PMLR.
  37. Rudin C. (2019). Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nature Machine Intelligence, 1(5), 206–215. doi : 10.1038/s42256-019-0048-x.
  38. Sia S. & Duh K. (2023). In-context learning as maintaining coherency: A study of on-the-fly machine translation using large language models.
  39. Somers H. (1999). Review article: Example-based machine translation. Machine Translation, 14(2), 113–157. doi : 10.1023/A:1008109312730.
  40. Attention is all you need. In Proceedings of the 31st International Conference on Neural Information Processing Systems, NIPS’17, p. 6000–6010, Red Hook, NY, USA: Curran Associates Inc.
  41. Prompting PaLM for translation: Assessing strategies and performance. In A. Rogers, J. Boyd-Graber & N. Okazaki, Éds., Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), p. 15406–15427, Toronto, Canada: Association for Computational Linguistics. doi : 10.18653/v1/2023.acl-long.859.
  42. Graph based translation memory for neural machine translation. Proceedings of the AAAI Conference on Artificial Intelligence, 33(01), 7297–7304. doi : 10.1609/aaai.v33i01.33017297.
  43. Xu J., Crego J. & Yvon F. (2023). Integrating translation memories into non-autoregressive machine translation. In A. Vlachos & I. Augenstein, Éds., Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, p. 1326–1338, Dubrovnik, Croatia: Association for Computational Linguistics. doi : 10.18653/v1/2023.eacl-main.96.
  44. Zhang B., Haddow B. & Birch A. (2023). Prompting large language model for machine translation: A case study. In Proceedings of the 40th International Conference on Machine Learning, ICML’23: JMLR.org.
  45. Guiding neural machine translation with retrieved translation pieces. In M. Walker, H. Ji & A. Stent, Éds., Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), p. 1325–1335, New Orleans, Louisiana: Association for Computational Linguistics. doi : 10.18653/v1/N18-1120.
  46. Towards a unified training for Levenshtein transformer. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), p. 1–5. doi : 10.1109/ICASSP49357.2023.10094646.
  47. Adaptive nearest neighbor machine translation. In C. Zong, F. Xia, W. Li & R. Navigli, Éds., 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), p. 368–374, Online: Association for Computational Linguistics. doi : 10.18653/v1/2021.acl-short.47.

Summary

No one has generated a summary of this paper yet.

Sign Up to Summarize

Paper to Video (Beta)

No one has generated a video about this paper yet.

Sign Up to Generate All Videos Subscribe on YouTube

Whiteboard

No one has generated a whiteboard explanation for this paper yet.

Sign Up to Generate

Open Problems

We haven't generated a list of open problems mentioned in this paper yet.

Generate Now

Continue Learning

We haven't generated follow-up questions for this paper yet.

Generate Now

Collections

Sign up for free to add this paper to one or more collections.

Sign Up

Tweets

Sign up for free to view the 1 tweet with 1 like about this paper.

Sign Up for Free