Elamur Research
☰
home / papers / 2023 / Prompting Large Language Model for Machine Translation: A Case Study

Prompting Large Language Model for Machine Translation: A Case Study

Written for the Research library (claude-opus-5-5) from the paper text, in our own words: the license "arXiv — non-exclusive license to distribute" does not allow redistributing the text itself. Original — arxiv.org/pdf/2301.07069

Summary

Prompting large language models works well for many NLP tasks, but its use for machine translation (MT) had not been studied systematically. The authors address this with a broad study on GLM-130B, a raw, non-finetuned 130B-parameter model trained on Chinese and English text, run in its INT4-quantized form. They cover English, German and Chinese, evaluating on FLORES, WMT21, Multi-Domain and the document-level PDC set with BLEU and COMET. The study examines four questions: how to design prompt templates, how to pick demonstration examples, whether monolingual data can help, and whether demonstrations transfer across domains, language pairs and from sentence-level to document-level translation.

Template choice strongly affects quality, and a minimal English template works best. Using more demonstrations, and better ones, improves output. Example features correlate with quality, but only weakly, so feature-based selection gives inconsistent gains. Raw monolingual examples harm translation, while pseudo-parallel examples created through zero-shot back- or forward-translation help. Transferred demonstrations outperform zero-shot. The work gives practical guidance for prompting-based MT and identifies where it still breaks down.

Key points

  • The best zero-shot template was the plain English form "[src]: X [tgt]:". It reached an average COMET of 38.78 over 6 language pairs, while most other templates scored negative. Templates written in German or Chinese gave mixed results.
  • Example features had Spearman correlations of only about 0.04–0.33 with prompting performance. These features were length, LMScore, MTScore and semantic similarity. Adding more demonstrations (K from 1 to 20) raised average COMET but increased memory use and per-token inference time.
  • Pseudo-parallel demonstrations built from monolingual data helped, and back-translation was stronger and more stable than forward-translation. Purely monolingual examples hurt quality.
  • Demonstrations from other domains, language pairs or sentence-level data beat the zero-shot baseline. However, an example that is optimal in one setting rarely stays optimal in another.
  • The remaining failure modes are:
    • copying the source,
    • mistranslating entities,
    • hallucination,
    • weak direct translation between non-English languages,
    • "prompt traps", where the prompt itself is hard to translate.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Other

Topics: Machine translation

Cited by

224 papers of our library cite it (of the 140673 whose bibliographies we read); 10 of them have a page here, 214 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.48550/arxiv.2301.07069 · arXiv 2301.07069

@misc{zhang2023prompting,
  title = {{Prompting Large Language Model for Machine Translation: A Case Study}},
  author = {Zhang, Biao and Haddow, Barry and Birch, Alexandra},
  year = {2023},
  doi = {10.48550/arxiv.2301.07069},
  eprint = {2301.07069},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.48550/arxiv.2301.07069}
}

References

46 references, 21 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sweta Agrawal, Chunting Zhou, Mike Lewis, Luke Zettlemoyer, and Marjan Ghazvininejad. 2022. In-context examples selection for machine translation. arXiv preprint arXiv:2212.02437. find our card
  2. Roee Aharoni and Yoav Goldberg. 2020. Unsupervised domain clusters in pretrained language models. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7747–7763, Online. Association for Computational Linguistics. find our card
  3. Farhad Akhbardeh, Arkady Arkhangorodsky, Magdalena Biesialska, Ondřej Bojar, Rajen Chatterjee, Vishrav Chaudhary, Marta R. Costa-jussa, Cristina España-Bonet, Angela Fan, Christian Federmann, Markus Freitag, Yvette Graham, Roman Grundkiewicz, Barry Haddow, Leonie Harter, Kenneth Heafield, Christoph… find our card
  4. Naveen Arivazhagan, Ankur Bapna, Orhan Firat, Dmitry Lepikhin, Melvin Johnson, Maxim Krikun, Mia Xu Chen, Yuan Cao, George Foster, Colin Cherry, Wolfgang Macherey, Zhifeng Chen, and Yonghui Wu. 2019. Massively multilingual neural machine translation in the wild: Findings and challenges. find our card
  5. Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel Ziegler, Jeffrey Wu, Clemens Winter, Chri… find
  6. Isaac Caswell, Ciprian Chelba, and David Grangier. 2019. Tagged back-translation. In Proceedings of the Fourth Conference on Machine Translation (Volume 1: Research Papers), pages 53–63, Florence, Italy. Association for Computational Linguistics. find our card
  7. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. 2022. Palm: Scaling language modeling with pathways. arXiv preprint arXiv:2204.02311. find
  8. Hyung Won Chung, Le Hou, Shayne Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, et al. 2022. Scaling instruction-finetuned language models. arXiv preprint arXiv:2210.11416. find
  9. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies,… find
  10. Yao Fu, Hao Peng, Ashish Sabharwal, Peter Clark, and Tushar Khot. 2022. Complexity-based prompting for multi-step reasoning. arXiv preprint arXiv:2210.00720. find our card
  11. Tianyu Gao, Adam Fisch, and Danqi Chen. 2021. Making pre-trained language models better few-shot learners. In 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), pa… find our card
  12. Xavier Garcia and Orhan Firat. 2022. Using natural language prompts for machine translation. arXiv preprint arXiv:2202.11822. — [arXiv:2202.11822](https://arxiv.org/abs/2202.11822) find
  13. Tanya Goyal, Junyi Jessy Li, and Greg Durrett. 2022. News summarization and evaluation in the era of gpt-3. arXiv preprint arXiv:2209.12356. find
  14. Jiatao Gu, Yong Wang, Kyunghyun Cho, and Victor O.K. Li. 2018. Search engine guided neural machine translation. In Proceedings of the Thirty-Second AAAI Conference on Artificial Intelligence and Thirtieth Innovative Applications of Artificial Intelligence Conference and Eighth AAAI Symposium on Educ… find
  15. Kevin Heffernan, Onur Çelebi, and Holger Schwenk. 2022. Bitext mining using distilled sentence representations for low-resource languages. arXiv preprint arXiv:2205.12654. find
  16. Melvin Johnson, Mike Schuster, Quoc Le, Maxim Krikun, Yonghui Wu, Zhifeng Chen, Nikhil Thorat, Fernanda Viégas, Martin Wattenberg, Greg Corrado, Macduff Hughes, and Jeffrey Dean. 2017. Google’s multilingual neural machine translation system: Enabling zero-shot translation. Transactions of the Associ… find
  17. Jared Kaplan, Sam McCandlish, Tom Henighan, Tom B Brown, Benjamin Chess, Rewon Child, Scott Gray, Alec Radford, Jeffrey Wu, and Dario Amodei. 2020. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361. find
  18. Yafu Li, Yongjing Yin, Jing Li, and Yue Zhang. 2022. Prompt-driven neural machine translation. In Findings of the Association for Computational Linguistics: ACL 2022, pages 2579–2590, Dublin, Ireland. Association for Computational Linguistics. find
  19. Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, et al. 2021. Few-shot learning with multilingual language models. arXiv preprint arXiv:2112.10668. find our card
  20. Jiachang Liu, Dinghan Shen, Yizhe Zhang, Bill Dolan, Lawrence Carin, and Weizhu Chen. 2022. What makes good in-context examples for GPT-3? In Proceedings of Deep Learning Inside Out (DeeLIO 2022): The 3rd Workshop on Knowledge Extraction and Integration for Deep Learning Architectures, pages 100–114… find our card
  21. Sewon Min, Xinxi Lyu, Ari Holtzman, Mikel Artetxe, Mike Lewis, Hannaneh Hajishirzi, and Luke Zettlemoyer. 2022. Rethinking the role of demonstrations: What makes in-context learning work? arXiv preprint arXiv:2202.12837. find our card
  22. Nikita Moghe, Tom Sherborne, Mark Steedman, and Alexandra Birch. 2022. Extrinsic evaluation of machine translation metrics. arXiv preprint arXiv:2212.10297. find
  23. NLLB Team, Marta R. Costa-jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, Anna Sun, Skyler Wang, Guillaume Wenzek, Al Youngblood, Bapi Akula, Loic Barrault, Gabriel Mejia Gonzalez, Prangthip Hansanti, John Hof… find
  24. Matt Post. 2018. A call for clarity in reporting BLEU scores. In Proceedings of the Third Conference on Machine Translation: Research Papers, pages 186– 191, Belgium, Brussels. Association for Computational Linguistics. find our card
  25. Ricardo Rei, Craig Stewart, Ana C Farinha, and Alon Lavie. 2020. COMET: A neural framework for MT evaluation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2685–2702, Online. Association for Computational Linguistics. find our card
  26. Holger Schwenk, Vishrav Chaudhary, Shuo Sun, Hongyu Gong, and Francisco Guzmán. 2021. Wiki-Matrix: Mining 135M parallel sentences in 1620 language pairs from Wikipedia. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages… find
  27. Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016a. Controlling politeness in neural machine translation via side constraints. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 35–40, San Dieg… find our card
  28. Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016b. Improving neural machine translation models with monolingual data. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 86–96, Berlin, Germany. Association for Computational… find
  29. Raphael Shu, Hideki Nakayama, and Kyunghyun Cho. 2019. Generating diverse translations with sentence codes. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 1823–1827, Florence, Italy. Association for Computational Linguistics. find
  30. Taylor Sorensen, Joshua Robinson, Christopher Rytting, Alexander Shaw, Kyle Rogers, Alexia Delorey, Mahmoud Khalil, Nancy Fulda, and David Wingate. 2022. An information-theoretic approach to prompt engineering without ground truth labels. In Proceedings of the 60th Annual Meeting of the Association… find
  31. Zewei Sun, Mingxuan Wang, Hao Zhou, Chengqi Zhao, Shujian Huang, Jiajun Chen, and Lei Li. 2022. Rethinking document-level neural machine translation. In Findings of the Association for Computational Linguistics: ACL 2022, pages 3537– 3548, Dublin, Ireland. Association for Computational Linguistics. find
  32. Zhixing Tan, Xiangwen Zhang, Shuo Wang, and Yang Liu. 2021. Msp: Multi-stage prompting for making pre-trained language models better translators. arXiv preprint arXiv:2110.06609. find
  33. David Vilar, Markus Freitag, Colin Cherry, Jiaming Luo, Viresh Ratnakar, and George Foster. 2022. Prompting palm for translation: Assessing strategies and performance. arXiv preprint arXiv:2211.09102. find
  34. Chengyu Wang, Jianing Wang, Minghui Qiu, Jun Huang, and Ming Gao. 2021. TransPrompt: Towards an automatic transferable prompting framework for few-shot text classification. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 2792–2802, Online and Punta Ca… find
  35. Jason Wei, Maarten Bosma, Vincent Zhao, Kelvin Guu, Adams Wei Yu, Brian Lester, Nan Du, Andrew M. Dai, and Quoc V Le. 2022a. Finetuned language models are zero-shot learners. In International Conference on Learning Representations. find
  36. Jason Wei, Yi Tay, Rishi Bommasani, Colin Raffel, Barret Zoph, Sebastian Borgeaud, Dani Yogatama, Maarten Bosma, Denny Zhou, Donald Metzler, et al. 2022b. Emergent abilities of large language models. arXiv preprint arXiv:2206.07682. find our card
  37. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Ed Chi, Quoc Le, and Denny Zhou. 2022c. Chain of thought prompting elicits reasoning in large language models. arXiv preprint arXiv:2201.11903. find our card
  38. Linting Xue, Noah Constant, Adam Roberts, Mihir Kale, Rami Al-Rfou, Aditya Siddhant, Aditya Barua, and Colin Raffel. 2021. mT5: A massively multilingual pre-trained text-to-text transformer. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Ling… find
  39. Aohan Zeng, Xiao Liu, Zhengxiao Du, Zihan Wang, Hanyu Lai, Ming Ding, Zhuoyi Yang, Yifan Xu, Wendi Zheng, Xiao Xia, Weng Lam Tam, Zixuan Ma, Yufei Xue, Jidong Zhai, Wenguang Chen, Peng Zhang, Yuxiao Dong, and Jie Tang. 2022. Glm-130b: An open bilingual pre-trained model. arXiv preprint arXiv:2210.02… find our card
  40. Biao Zhang, Philip Williams, Ivan Titov, and Rico Sennrich. 2020. Improving massively multilingual neural machine translation and zero-shot translation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1628–1639, Online. Association for Computational… find our card
  41. Jiajun Zhang and Chengqing Zong. 2016. Exploiting source-side monolingual data in neural machine translation. In Proceedings of the 2016 Conference on Empirical Methods in Natural Language Processing, pages 1535–1545. find our card
  42. Jingyi Zhang, Masao Utiyama, Eiichro Sumita, Graham Neubig, and Satoshi Nakamura. 2018. Guiding neural machine translation with retrieved translation pieces. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technolog… find
  43. Susan Zhang, Stephen Roller, Naman Goyal, Mikel Artetxe, Moya Chen, Shuohui Chen, Christopher Dewan, Mona Diab, Xian Li, Xi Victoria Lin, et al. 2022a. Opt: Open pre-trained transformer language models. arXiv preprint arXiv:2205.01068. find
  44. Yiming Zhang, Shi Feng, and Chenhao Tan. 2022b. Active example selection for in-context learning. arXiv preprint arXiv:2211.04486. find our card
  45. Zihao Zhao, Eric Wallace, Shi Feng, Dan Klein, and Sameer Singh. 2021. Calibrate before use: Improving few-shot performance of language models. In Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proceedings of Machine Learning Research, pages 12697–12706. PMLR. find our card
  46. Denny Zhou, Nathanael Schärli, Le Hou, Jason Wei, Nathan Scales, Xuezhi Wang, Dale Schuurmans, Olivier Bousquet, Quoc Le, and Ed Chi. 2022. Least-to-most prompting enables complex reasoning in large language models. arXiv preprint arXiv:2205.10625. find our card