Improving Machine Translation Formality with Large Language Models
Source abstract
Abstract
Preserving formal style in neural machine translation (NMT) is essential, yet often overlooked as an optimization objective of the training processes. This oversight can lead to translations that, though accurate, lack formal... | Find, read and cite all the research you need on Tech Science Press
Source: openalex_inverted_index · Quality: baseline
Research context
OpenAlex topics: Natural Language Processing Techniques · Topic Modeling
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
References (38)
Show source-provided references
1. Controlling neural machine translation formality with synthetic supervision
2. Improving stylized neural machine translation with iterative dual knowledge transfer
3. Y. Wang, Z. Sun, S. Cheng, W. Zheng, and M. Wang, “Controlling styles in neural machine translation with activation prompt,” 2022, arXiv:2212.08909.10.18653/v1/2023.findings-acl.163 ↗
4. Tutorial on maximum likelihood estimation10.1016/S0022-2496(02)00028-7 ↗
5. A study of style in machine translation: Controlling the formality of machine translation output
6. T. Kocmi and C. Federmann, “Large language models are state-of-the-art evaluators of translation quality,” 2023, arXiv:2302.14520.
7. C. Wang et al., “Learning evaluation models from large language models for sequence generation,” 2023, arXiv:2308.04386.
8. Simple statistical gradient-following algorithms for connectionist reinforcement learning10.1023/A:1022672621406 ↗
9. J. Schulman, F. Wolski, P. Dhariwal, A. Radford, and O. Klimov, “Proximal policy optimization algorithms,” 2017, arXiv:1707.06347.
10. Personalized machine translation: Preserving original author traits
11. Extreme adaptation for personalized neural machine translation
12. Shakespearizing modern language using copy-enriched sequence to sequence models
13. Parallel data augmentation for formality style transfer
14. Formality style transfer with shared latent space
15. Thank you BART! rewarding pre-trained models improves formality style transfer
16. B. Li et al., “Deliberate then generate: Enhanced prompting framework for text generation,” 2023, arXiv:2305.19835.
17. H. Lai, A. Toral, and M. Nissim, “Multidimensional evaluation for text style transfer using ChatGPT,” 2023, arXiv:2304.13462.
18. BLEU: A method for automatic evaluation of machine translation
19. Minimum risk training for neural machine translation
20. Revisiting the weaknesses of reinforcement learning for neural machine translation
21. D. Donato, L. Yu, W. Ling, and C. Dyer, “Mad for robust reinforcement learning in machine translation,” ArXiv preprint, 2022.
22. Brown · Advances in Neural Information Processing Systems · Dec. 6–12, 2020 · 33 · 1877
23. ESRL: Efficient sampling-based reinforcement learning for sequence generation
24. N. Ho, L. Schmid, and S. -Y. Yun, “Large language models are reasoning teachers,” 2022, arXiv:2212.10071.
25. Wang · Findings of the Association for Computational Linguistics: EMNLP 2022 · 2022 · 6232
26. Language models are unsupervised multitask learners
27. Y. Liu et al., “RoBERTa: A robustly optimized BERT pretraining approach,” 2019, arXiv:1907.11692.
28. Vaswani · Advances in Neural Information Processing Systems · Dec. 4–9, 2017 · 5998
29. C. Hu et al., “RankNAS: Efficient neural architecture search by pairwise ranking,” 2021, arXiv:2109.07383.10.18653/v1/2021.emnlp-main.191 ↗
30. Dear sir or madam, may I introduce the GYAFC dataset: Corpus, benchmarks and metrics for formality style transfer
31. T. Xiao and J. Zhu, “Introduction to transformers: An NLP perspective,” 2023, arXiv:2311.17633.
32. Christiano · Advances in Neural Information Processing Systems · Dec. 4–9, 2017 · 4299
33. J. Wang et al., “Is ChatGPT a good NLG evaluator? A preliminary study,” 2023, arXiv:2303.04048.10.18653/v1/2023.newsum-1.1 ↗
34. COMET-22: Unbabel-IST, 2022 submission for the metrics shared task
35. Convolutional neural networks for sentence classification
36. GLM: General language model pretraining with autoregressive blank infilling
37. H. Touvron et al., “LLaMA: Open and efficient foundation language models,” 2023, arXiv:2302.13971.
38. The curious case of neural text degeneration