Elamur Research
☰
home / papers / 2023 / The Devil Is in the Errors: Leveraging Large Language Models for Fine-grained Machine Translation Evaluation

The Devil Is in the Errors: Leveraging Large Language Models for Fine-grained Machine Translation Evaluation

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2023.wmt-1.100.pdf

Summary

Automatic MT metrics, even strong learned ones, usually output a single number that does not show what went wrong in a translation. Human MQM annotation, by contrast, marks individual error spans with a category and severity. The authors test whether large language models, given only a few labelled examples, can produce MQM-style feedback. They first study PaLM and PaLM-2 models as score predictors, with and without references. They compare zero-shot prompting, in-context examples and finetuning on human judgments, and they include low-resource language pairs. They then introduce AutoMQM, a prompt that asks the model to list erroneous spans and label each one by type and severity. The final score is computed from these labels using MQM weights instead of being requested from the model. On WMT22, plain score prompting already ranks systems at state-of-the-art level but trails learned metrics per segment unless the model is finetuned. AutoMQM improves segment-level correlation for the larger PaLM-2 UNICORN, produces more realistic score distributions and gives interpretable error spans. This suggests LLMs can provide explainable feedback for evaluating generated text.

Key points

  • Score prediction uses the GEMBA-SQM prompt (0–100 scale). LLMs used this way are top system-level evaluators but correlate weakly at segment level. Finetuning on human judgments closes most of this gap, especially for smaller models.
  • The study is the first to test LLM-based MT evaluation on low-resource pairs: EN-KK, EN-GU, KK-EN and GU-EN.
  • With AutoMQM and a reference, PaLM-2 UNICORN reaches 87.6% system-level accuracy, compared with 81.1% for MetricX-XXL. Its segment-level Pearson is 0.432 (EN-DE) and 0.442 (ZH-EN), up from 0.401 and 0.349 with score prediction. The smaller BISON gains little, which points to scale mattering.
  • At span level, PaLM-2 models recover over half of the words inside major errors. Their MCC is comparable to the supervised word-level model COMET-WL, and their major-error recall is much higher (e.g., 0.628 vs 0.250 on EN-DE for UNICORN with reference).
  • The models tend to over-predict errors: UNICORN spans average about 5 words, against about 2 in gold annotations, which lowers precision. Baselines also include MaTESe and COMET-QE. Model outputs are released in the google-research GitHub repository.

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: Machine translation

Topics: Machine translation

Cited by

123 papers of our library cite it (of the 140673 whose bibliographies we read); 10 of them have a page here, 113 not yet.

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2023.wmt-1.100

@inproceedings{fernandes2023devil,
  title = {{The Devil Is in the Errors: Leveraging Large Language Models for Fine-grained Machine Translation Evaluation}},
  author = {Fernandes, Patrick and Deutsch, Daniel and Finkelstein, Mara and Riley, Parker and Martins, André and Neubig, Graham and Garg, Ankush and Clark, Jonathan and Freitag, Markus and Firat, Orhan},
  booktitle = {Proceedings of the Eighth Conference on Machine Translation},
  year = {2023},
  doi = {10.18653/v1/2023.wmt-1.100},
  url = {https://doi.org/10.18653/v1/2023.wmt-1.100}
}

References

35 references, 14 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Rohan Anil, Andrew M. Dai, Orhan Firat, Melvin Johnson, Dmitry Lepikhin, Alexandre Passos, Siamak Shakeri, Emanuel Taropa, Paige Bailey, Zhifeng Chen, Eric Chu, Jonathan H. Clark, Laurent El Shafey, Yanping Huang, Kathy Meier-Hellstern, Gaurav Mishra, Erica Moreira, Mark Omernick, Kevin Robinson, Sebastian Ruder, Yi Tay, Kefan Xiao, Yuanzhong Xu, Yujing Zhang, Gustavo Hernandez Abrego, Junwhan Ahn, Jacob Austin, Paul Barham, Jan Botha, James Bradbury, Siddhartha Brahma, Kevin Brooks, Michele Catasta, Yong Cheng, Colin Cherry, Christopher A. Choquette-Choo, Aakanksha Chowdhery, Clément Crepy, Shachi Dave, Mostafa Dehghani, Sunipa Dev, Jacob Devlin, Mark Díaz, Nan Du, Ethan Dyer, Vlad Feinberg, Fangxiaoyu Feng, Vlad Fienber, Markus Freitag, Xavier Garcia, Sebastian Gehrmann, Lucas Gonzalez, Guy Gur-Ari, Steven Hand, Hadi Hashemi, Le Hou, Joshua Howland, Andrea Hu, Jeffrey Hui, Jeremy Hurwitz, Michael Isard, Abe Ittycheriah, Matthew Jagielski, Wenhao Jia, Kathleen Kenealy, Maxim Krikun, Sneha Kudugunta, Chang Lan, Katherine Lee, Benjamin Lee, Eric Li, Music Li, Wei Li, YaGuang Li, Jian Li, Hyeontaek Lim, Hanzhao Lin, Zhongtao Liu, Frederick Liu, Marcello Maggioni, Aroma Mahendru, Joshua Maynez, Vedant Misra, Maysam Moussalem, Zachary Nado, John Nham, Eric Ni, Andrew Nystrom, Alicia Parrish, Marie Pellat, Martin Polacek, Alex Polozov, Reiner Pope, Siyuan Qiao, Emily Reif, Bryan Richter, Parker Riley, Alex Castro Ros, Aurko Roy, Brennan Saeta, Rajkumar Samuel, Renee Shelby, Ambrose Slone, Daniel Smilkov, David R. So, Daniel Sohn, Simon Tokumine, Dasha Valter, Vijay Vasudevan, Kiran Vodrahalli, Xuezhi Wang, Pidong Wang, Zirui Wang, Tao Wang, John Wieting, Yuhuai Wu, Kelvin Xu, Yunhan Xu, Linting Xue, Pengcheng Yin, Jiahui Yu, Qiao Zhang, Steven Zheng, Ce Zheng, Weikang Zhou, Denny Zhou, Slav Petrov, and Yonghui Wu. 2023. Palm 2 technical report. find our card
  2. 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, Chris Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. 2020. Language models are few-shot learners. In Advances in Neural Information Processing Systems, volume 33, pages 1877–1901. Curran Associates, Inc. find
  3. Chris Callison-Burch, Philipp Koehn, Christof Monz, Josh Schroeder, and Cameron Shaw Fordyce, editors. 2008. Proceedings of the Third Workshop on Statistical Machine Translation. Association for Computational Linguistics, Columbus, Ohio. find
  4. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, Parker Schuh, Kensen Shi, Sasha Tsvyashchenko, Joshua Maynez, Abhishek Rao, Parker Barnes, Yi Tay, Noam Shazeer, Vinodkumar Prabhakaran, Emily Reif, Nan Du, Ben Hutchinson, Reiner Pope, James Bradbury, Jacob Austin, Michael Isard, Guy Gur-Ari, Pengcheng Yin, Toju Duke, Anselm Levskaya, Sanjay Ghemawat, Sunipa Dev, Henryk Michalewski, Xavier Garcia, Vedant Misra, Kevin Robinson, Liam Fedus, Denny Zhou, Daphne Ippolito, David Luan, Hyeontaek Lim, Barret Zoph, Alexander Spiridonov, Ryan Sepassi, David Dohan, Shivani Agrawal, Mark Omernick, Andrew M. Dai, Thanumalayan Sankaranarayana Pillai, Marie Pellat, Aitor Lewkowycz, Erica Moreira, Rewon Child, Oleksandr Polozov, Katherine Lee, Zongwei Zhou, Xuezhi Wang, Brennan Saeta, Mark Diaz, Orhan Firat, Michele Catasta, Jason Wei, Kathy Meier-Hellstern, Douglas Eck, Jeff Dean, Slav Petrov, and Noah Fiedel. 2022. Palm: Scaling language modeling with pathways. find
  5. Alexis Conneau, Kartikay Khandelwal, Naman Goyal, Vishrav Chaudhary, Guillaume Wenzek, Francisco Guzmán, Edouard Grave, Myle Ott, Luke Zettlemoyer, and Veselin Stoyanov. 2020. Unsupervised cross-lingual representation learning at scale. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 8440– 8451, Online. Association for Computational Linguistics. ↗
  6. Daniel Deutsch, George Foster, and Markus Freitag. 2023. Ties Matter: Modifying Kendall’s Tau for Modern Metric Meta-Evaluation. find
  7. Patrick Fernandes, Aman Madaan, Emmy Liu, António Farinhas, Pedro Henrique Martins, Amanda Bertsch, José G. C. de Souza, Shuyan Zhou, Tongshuang Wu, Graham Neubig, and André F. T. Martins. 2023. Bridging the gap: A survey on integrating (human) feedback for natural language generation. find
  8. Erick Fonseca, Lisa Yankovskaya, Andrà c F. T. Martins, Mark Fishel, and Christian Federmann. 2019. Findings of the wmt 2019 shared tasks on quality estimation. In Proceedings of the Fourth Conference on Machine Translation (Volume 3: Shared Task Papers, Day 2), pages 1–12, Florence, Italy. Association for Computational Linguistics. find our card
  9. Markus Freitag, George Foster, David Grangier, Viresh Ratnakar, Qijun Tan, and Wolfgang Macherey. 2021a. Experts, errors, and context: A large-scale study of human evaluation for machine translation. Transactions of the Association for Computational Linguistics, 9:1460–1474. find our card
  10. Markus Freitag, Ricardo Rei, Nitika Mathur, Chi-kiu Lo, Craig Stewart, Eleftherios Avramidis, Tom Kocmi, George Foster, Alon Lavie, and André F. T. Martins. 2022. Results of WMT22 metrics shared task: Stop using BLEU – neural metrics are better and more robust. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 46–68, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find our card
  11. Markus Freitag, Ricardo Rei, Nitika Mathur, Chi-kiu Lo, Craig Stewart, George Foster, Alon Lavie, and Ondřej Bojar. 2021b. Results of the WMT21 metrics shared task: Evaluating metrics with expert-based human evaluations on TED and news domain. In Proceedings of the Sixth Conference on Machine Translation, pages 733–774, Online. Association for Computational Linguistics. find our card
  12. Luyu Gao, Aman Madaan, Shuyan Zhou, Uri Alon, Pengfei Liu, Yiming Yang, Jamie Callan, and Graham Neubig. 2022. Pal: Program-aided language models. arXiv preprint arXiv:2211.10435. ↗
  13. Sameer Jain, Vaishakh Keshava, Swarnashree Mysore Sathyendra, Patrick Fernandes, Pengfei Liu, Graham Neubig, and Chunting Zhou. 2023. Multi-dimensional evaluation of text summarization with in-context learning. find
  14. Tom Kocmi, Rachel Bawden, Ondřej Bojar, Anton Dvorkovich, Christian Federmann, Mark Fishel, Thamme Gowda, Yvette Graham, Roman Grundkiewicz, Barry Haddow, Rebecca Knowles, Philipp Koehn, Christof Monz, Makoto Morishita, Masaaki Nagata, Toshiaki Nakazawa, Michal Novák, Martin Popel, and Maja Popović. 2022. Findings of the 2022 conference on machine translation (WMT22). In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 1–45, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find
  15. Tom Kocmi and Christian Federmann. 2023. Large language models are state-of-the-art evaluators of translation quality. find
  16. Tom Kocmi, Christian Federmann, Roman Grundkiewicz, Marcin Junczys-Dowmunt, Hitokazu Matsushita, and Arul Menezes. 2021. To ship or not to ship: An extensive evaluation of automatic metrics for machine translation. In Proceedings of the Sixth Conference on Machine Translation, pages 478–494, Online. Association for Computational Linguistics. find
  17. Pengfei Liu, Weizhe Yuan, Jinlan Fu, Zhengbao Jiang, Hiroaki Hayashi, and Graham Neubig. 2023a. Pre-train, prompt, and predict: A systematic survey of prompting methods in natural language processing. ACM Comput. Surv., 55(9). find our card
  18. Yang Liu, Dan Iter, Yichong Xu, Shuohang Wang, Ruochen Xu, and Chenguang Zhu. 2023b. G-eval: Nlg evaluation using gpt-4 with better human alignment. find
  19. Arle Lommel, Hans Uszkoreit, and Aljoscha Burchardt. 2014. Multidimensional quality metrics (MQM): A framework for declaring and describing translation quality metrics. Revista Tradumàtica: tecnologies de la traducció. find
  20. Qingyu Lu, Baopu Qiu, Liang Ding, Liping Xie, and Dacheng Tao. 2023. Error analysis prompting enables human-like translation evaluation in large language models: A case study on chatgpt. arXiv preprint. find
  21. Qingsong Ma, Johnny Wei, Ondřej Bojar, and Yvette Graham. 2019. Results of the WMT19 metrics shared task: Segment-level and strong MT systems pose big challenges. In Proceedings of the Fourth Conference on Machine Translation (Volume 2: Shared Task Papers, Day 1), pages 62–90, Florence, Italy. Association for Computational Linguistics. find our card
  22. OpenAI. 2023. Gpt-4 technical report. find our card
  23. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics, pages 311–318, Philadelphia, Pennsylvania, USA. Association for Computational Linguistics. ↗ our card
  24. Stefano Perrella, Lorenzo Proietti, Alessandro Scirè, Niccolò Campolungo, and Roberto Navigli. 2022. MaTESe: Machine translation evaluation as a sequence tagging problem. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 569–577, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find
  25. Yiwei Qin, Weizhe Yuan, Graham Neubig, and Pengfei Liu. 2022. T5score: Discriminative fine-tuning of generative evaluation metrics. ArXiv, abs/2212.05726. ↗
  26. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res., 21(1). find
  27. Ricardo Rei, José G. C. de Souza, Duarte Alves, Chrysoula Zerva, Ana C Farinha, Taisiya Glushkova, Alon Lavie, Luisa Coheur, and André F. T. Martins. 2022a. COMET-22: Unbabel-IST 2022 submission for the metrics shared task. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 578–585, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find our card
  28. Ricardo Rei, Marcos Treviso, Nuno M. Guerreiro, Chrysoula Zerva, Ana C Farinha, Christine Maroti, José G. C. de Souza, Taisiya Glushkova, Duarte Alves, Luisa Coheur, Alon Lavie, and André F. T. Martins. 2022b. CometKiwi: IST-unbabel 2022 submission for the quality estimation shared task. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 634–645, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find our card
  29. Thibault Sellam, Dipanjan Das, and Ankur Parikh. 2020. BLEURT: Learning robust metrics for text generation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7881–7892, Online. Association for Computational Linguistics. find
  30. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, brian ichter, Fei Xia, Ed H. Chi, Quoc V Le, and Denny Zhou. 2022. Chain of thought prompting elicits reasoning in large language models. In Advances in Neural Information Processing Systems. find our card
  31. Zeqiu Wu, Yushi Hu, Weijia Shi, Nouha Dziri, Alane Suhr, Prithviraj Ammanabrolu, Noah A. Smith, Mari Ostendorf, and Hannaneh Hajishirzi. 2023. Fine-grained human feedback gives better rewards for language model training. find our card
  32. Wenda Xu, Danqing Wang, Liangming Pan, Zhenqiao Song, Markus Freitag, William Yang Wang, and Lei Li. 2023. Instructscore: Towards explainable text generation evaluation with automatic feedback. find
  33. 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 Linguistics: Human Language Technologies, pages 483–498, Online. Association for Computational Linguistics. ↗
  34. Chrysoula Zerva, Frédéric Blain, Ricardo Rei, Piyawat Lertvittayakumjorn, José G. C. de Souza, Steffen Eger, Diptesh Kanojia, Duarte Alves, Constantin Orăsan, Marina Fomicheva, André F. T. Martins, and Lucia Specia. 2022. Findings of the WMT 2022 shared task on quality estimation. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 69–99, Abu Dhabi, United Arab Emirates (Hybrid). Association for Computational Linguistics. find our card
  35. Chrysoula Zerva, Daan van Stigt, Ricardo Rei, Ana C Farinha, Pedro Ramos, José G. C. de Souza, Taisiya Glushkova, Miguel Vera, Fabio Kepler, and André F. T. Martins. 2021. IST-unbabel 2021 submission for the quality estimation shared task. In Proceedings of the Sixth Conference on Machine Translation, pages 961–972, Online. Association for Computational Linguistics. find