Elamur Research⌕
☰
home / papers / 2026 / Beyond Simple Term Injection: Reasoning Models for Legal Translation in a Non-Dominant Language Variety

Beyond Simple Term Injection: Reasoning Models for Legal Translation in a Non-Dominant Language Variety

Summary

Recent WMT terminology results suggest that LLMs have almost solved sentence-level injection of glossary terms. This paper tests whether that claim holds in harder conditions. The setting is Italian-to-South Tyrolean German legal translation. South Tyrolean German is a non-dominant, low-resource variety, and its legal terms differ from those used in Germany, Austria and Switzerland. The authors use the curated LegISTyr test set, which covers three challenges: plain term injection, localising abbreviated forms, and picking the right target equivalent for homonymous source terms. They compare reasoning models that use test-time scaling with standard LLMs, translation-specialised LLMs and fine-tuned NMT systems. Outputs are scored on term accuracy and MetricX-XL fluency, and the authors also read the reasoning traces manually. Reasoning barely helps with plain injection but clearly improves homonym disambiguation. However, the traces rarely take the language variety into account and justify choices inconsistently. When no terminology is supplied, even strong models miss the local terms, while small NMT models trained on in-domain parallel data remain competitive. The findings argue for variety-aware termbases and for reasoning data built specifically for translation.

Key points

  • LegISTyr is divided into three parts: 1250 instances test straightforward insertion of terms, 290 target abbreviated forms, and 250 deal with homonyms. Each source–target term pair appears in 5 sentences from the same legal subdomain. Reference terminology comes from bistro, which holds about 14,000 Italian legal concepts with validated South Tyrolean German terms.
  • Simple injection is nearly saturated when terms are given in the prompt. Most LLMs reach roughly 92–97% accuracy, for example GPT 4.1 at 97.52% and GPT 5.1 at 97.12%, and enabling thinking gives no consistent gain.
  • On Homonyms, reasoning helps substantially:
    • Gemini 2.5 Flash rises from 70.00% without thinking to 87.60% with about 869 reasoning tokens.
    • Qwen3-Plus rises from 75.60% to 83.60% with about 1317 tokens.
    • GPT 5.1 reaches the best score, 88.40%.
  • Without terms in the prompt, accuracy on Simple terms collapses for LLMs: 36.64% for Gemini 2.5 Flash and 14.88% for Qwen 8b. The fine-tuned NMT models Opus it-de and Madlad-400 3b reach 49.36% and 55.84%.
  • Prompting in German instead of English did not raise homonym accuracy for Gemini 2.5 Flash, GPT 5.1 or Qwen3-Plus. The authors propose two directions for future work:
    • collecting reasoning chains based on human translation;
    • treating language varieties as a core design principle when building termbases.

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: Linguistics & languages · Machine translation · Law

Topics: Machine translation

Cite

The paper itself, not this page. BibTeX from the library's record of the work.

@inproceedings{natale2026beyond,
  title = {{Beyond Simple Term Injection: Reasoning Models for Legal Translation in a Non-Dominant Language Variety}},
  author = {Natale, Paolo Di and Chiocchetti, Elena and Alber, Marlies and Stemle, Egon W.},
  booktitle = {Proceedings of the 26th Annual Conference of the European Association for Machine Translation (Volume 1)},
  year = {2026},
  url = {https://aclanthology.org/2026.eamt-1.23/}
}

References

90 references, 20 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ammon, Ulrich, Hans Bickel, and Alexandra Nicole Lenz. 2016. Variantenwörterbuch des Deutschen. Die Standardsprache in Österreich, der Schweiz, Deutschland, Liechtenstein, Luxemburg, Ostbelgien und Südtirol sowie Rumänien, Namibia und Mennon-itensiedlungen. de Gruyter, Berlin, 2 edition. find
  2. Astatinfo. 2024. Ergebnisse Sprachgruppenzählung - 2024 / Risultati Censimento linguistico - 2024. Technical Report 56, Astat – Landesinstitut für Statistik / Istituto provinciale di statistica, Bozen/Bolzano. https://assets-eu-01.kc-usercontent.com/ b5376750-8076-01cf-17d2-d343e29778a7/ 5deec178-b2a3-4e2d-8795-d37635c7e0f7/ pressnote\_1160209\_mit56\_2024.pdf. ↗
  3. Bai, Yifan, Yiping Bao, Y. Charles, Cheng Chen, Guanduo Chen, Haiting Chen, Huarong Chen, Jiahao Chen, Ningxin Chen, Ruijue Chen, Yanru Chen, Yuankun Chen, Yutian Chen, Zhuofu Chen, Jialei Cui, Hao Ding, Mengnan Dong, Angang Du, Chenzhuang Du, Dikang Du, Yulun Du, Yu Fan, Yichen Feng, Kelin Fu, Bofei Gao, Chenxiao Gao, Hongcheng Gao, Peizhong Gao, Tong Gao, Yuyao Ge, Shangyi Geng, Qizheng Gu, Xinran Gu, Longyu Guan, Haiqing Guo, Jianhang Guo, Xiaoru Hao, Tianhong He, Weiran He, Wenyang He, Yunjia He, Chao Hong, Hao Hu, Yangyang Hu, Zhenxing Hu, Weixiao Huang, Zhiqi Huang, Zihao Huang, Tao Jiang, Zhejun Jiang, Xinyi Jin, Yongsheng Kang, Guokun Lai, Cheng Li, Fang Li, Haoyang Li, Ming Li, Wentao Li, Yang Li, Yanhao Li, Yiwei Li, Zhaowei Li, Zheming Li, Hongzhan Lin, Xiaohan Lin, Zongyu Lin, Chengyin Liu, Chenyu Liu, Hongzhang Liu, Jingyuan Liu, Junqi Liu, Liang Liu, and Shaowei Liu. 2026. Kimi K2: Open Agentic Intelligence. Arxiv:2507.20534. ↗ our card
  4. Bernardini, Silvia. 2002. Think-aloud protocols in translation research: Achievements, limits, future prospects. Target, 13:241–263, 06. ↗
  5. Biel, Lucja and Hendrik Kockaert. 2023. Introduction. legal terminology. In Biel, Lucja and Hendrik Kockaert, editors, Handbook of Terminology. Legal Terminology, volume 3, page 1–14. John Benjamins, Amsterdam / Philadelphia. find
  6. Biel, Lucja. 2023. Variation of legal terms in monolingual and multilingual contexts. In Biel, Lucja and Hendrik J. Kockaert, editors, Handbook of Terminology. Legal Terminology, volume 3, page 90–123. John Benjamins, Amsterdam/Philadelphia. Bogoychev, Nikolay and Pinzhen Chen. Terminology-aware translation with constrained decoding and large language model prompting. In Koehn, Philipp, Barry Haddow, Tom Kocmi, and Christof Monz, editors, Proceedings of the Eighth Conference on Machine Translation, pages 890–896, Singapore, December. Association for Computational Linguistics. ↗
  7. Briakou, Eleftheria, Jiaming Luo, Colin Cherry, and Markus Freitag. 2024. Translating step-by-step: Decomposing the translation process for improved translation quality of long-form texts. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Ninth Conference on Machine Translation, pages 1301–1317, Miami, Florida, USA, November. Association for Computational Linguistics. find our card
  8. Briva-Iglesias, Vicent. 2025. Are AI agents the new machine translation frontier? Challenges and opportunities of single- and multi-agent systems for multilingual digital communication. In Bouillon, Pierrette, Johanna Gerlach, Sabrina Girletti, Lise Volkart, Raphael Rubino, Rico Sennrich, Ana C. Farinha, Marco Gaido, Joke Daems, Dorothy Kenny, Helena Moniz, and Sara Szoc, editors, Proceedings of Machine Translation Summit XX: Volume 1, pages 365–377, Geneva, Switzerland, June. European Association for Machine Translation. find
  9. Canfora, Carmen and Angelika Ottmann. 2018. Of ostriches, pyramids, and Swiss cheese: Risks in safety-critical translations. Translation Spaces, 7(2):167–201. ↗
  10. Castilho, Sheila, Zoe Fitzsimmons, Claire Holton, and Aoife Mc Donagh. 2025. Synthetic fluency: Hallucinations, confabulations, and the creation of IrishWords in LLM-generated translations. In Bouillon, Pierrette, Johanna Gerlach, Sabrina Girletti, Lise Volkart, Raphael Rubino, Rico Sennrich, Ana C. Farinha, Marco Gaido, Joke Daems, Dorothy Kenny, Helena Moniz, and Sara Szoc, editors, Proceedings of Machine Translation Summit XX: Volume 1, pages 287–299, Geneva, Switzerland, June. European Association for Machine Translation. find
  11. Chen, Andong, Yuchen Song, Wenxin Zhu, Kehai Chen, Muyun Yang, Tiejun Zhao, and Min zhang. 2025. Evaluating o1-Like LLMs: Unlocking reasoning for translation through comprehensive analysis. Arxiv:2502.11544. d’Aliberti, Liv G. and Manoel Horta Ribeiro. 2026. The Illusion of Insight in Reasoning Models. Arxiv:2601.00514. ↗
  12. Dhuliawala, Shehzaad, Ilia Kulikov, Ping Yu, Asli Celikyilmaz, Jason Weston, Sainbayar Sukhbaatar, and Jack Lanchantin. 2024. Adaptive Decoding via Latent Preference Optimization. CoRR, abs/2411.09661. ↗
  13. Di Natale, Paolo, Elena Chiocchetti, and Egon W. Stemle. 2025a. Meta-evaluation of automatic machine translation metrics between Italian and a minor language variety of German. In Bosco, Cristina, Elisabetta Jezek, Marco Polignano, and Manuela Sanguinetti, edi-2023. tors, Proceedings of the Eleventh Italian Conference on Computational Linguistics (CLiC-it 2025), pages 371– 383, Cagliari, Italy, September. CEUR Workshop Proceedings. find
  14. Di Natale, Paolo, Egon W. Stemle, Elena Chiocchetti, Marlies Alber, Natascia Ralli, Isabella Stanizzi, and Elena Benini. 2025b. The LegISTyr test set: Investigating off-the-shelf instruction-tuned LLMs for terminology-constrained translation in a low-resource language variety. In Gkirtzou, Katerina, Slavko Žitnik, Jorge Gracia, Dagmar Gromann, Maria Pia di Buono, Johanna Monti, and Maxim Ionov, editors, Proceedings of the 5th Conference on Language, Data and Knowledge: TermTrends 2025, pages 1–15, Naples, Italy, September. Unior Press. find
  15. Di Nunzio, Giorgio Maria. 2025. Terminology-Augmented Generation (TAG): Foundations, Use Cases, and Evaluation Paths. Journal of Digital Terminology and Lexicography, 1(1). ↗
  16. Dong, Qingxiu, Lei Li, Damai Dai, Ce Zheng, Jingyuan Ma, Rui Li, Heming Xia, Jingjing Xu, Zhiyong Wu, Baobao Chang, Xu Sun, Lei Li, and Zhifang Sui. 2024. A survey on in-context learning. In Al-Onaizan, Yaser, Mohit Bansal, and Yun-Nung Chen, editors, Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 1107–1128, Miami, Florida, USA, November. Association for Computational Linguistics. find our card
  17. Doren, Madison Van, Casey Ford, Jennifer Barajas, and Cory Holland. 2026. ”Be My Cheese?”: Cultural Nuance Benchmarking for Machine Translation in Multilingual LLMs. Arxiv:2602.04729. ↗
  18. Gehrmann, Sebastian, Elizabeth Clark, and Thibault Sellam. 2023. Repairing the Cracked Foundation: A Survey of Obstacles in Evaluation Practices for Generated Text. J. Artif. Int. Res., 77, June. ↗
  19. Gema, Aryo Pradipta, Alexander Hägele, Runjin Chen, Andy Arditi, Jacob Goldman-Wetzler, Kit Fraser-Taliente, Henry Sleight, Linda Petrini, Julian Michael, Beatrice Alex, Pasquale Minervini, Yanda Chen, Joe Benton, and Ethan Perez. 2025. Inverse Scaling in Test-Time Compute. Arxiv:2507.14417. ↗
  20. Gobjila, Natalia. 2024. The problem of abbreviations in specialized translation. Studia Universitatis Moldaviae (Seria Ştiinţe Umanistice), (10):226–230. ↗
  21. Guo, Daya, Dejian Yang, Haowei Zhang, Junxiao Song, Peiyi Wang, Qihao Zhu, Runxin Xu, Ruoyu Zhang, Shirong Ma, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, Aixin Liu, Bing Xue, Bingxuan Wang, Bochao Wu, Bei Feng, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chong Ruan, Damai Dai, and Deli Chen. 2025. DeepSeek-R1 incentivizes reasoning in LLMs through reinforcement learning. Nature, 645(8081):633–638, September. ↗ our card
  22. Huan, Maggie, Yuetai Li, Tuney Zheng, Xiaoyu Xu, Seungone Kim, Minxin Du, Radha Poovendran, Graham Neubig, and Xiang Yue. 2025. Does Math Reasoning Improve General LLM capabilities? Understanding Transferability of LLM Reasoning. arXiv:2507.00432. ↗
  23. ISO 1087. 2019. Terminology work and terminology science — Vocabulary. find
  24. Juraska, Juraj, Daniel Deutsch, Mara Finkelstein, and Markus Freitag. 2024. MetricX-24: The Google submission to the WMT 2024 metrics shared task. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Ninth Conference on Machine Translation, pages 492–504, Miami, Florida, USA, November. Association for Computational Linguistics. find our card
  25. Kang, Deokhyung, Seonjeong Hwang, Daehui Kim, Hyounghun Kim, and Gary Geunbae Lee. 2026. Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models? Arxiv:2510.27269. ↗
  26. Khemakhem, Mohamed, Cristina Valentini, Natascia Ralli, Sérgio Barros, Georg Löckinger, Federica Vezzani, Ana Salgado, Zhenling Zhang, Sabine Mahr, Sara Carvalho, Klaus Fleischmann, and Rute Costa. 2025. Terminology management meets AI: The ISO/TC 37/SC 3/WG 6 initiative. In Gkirtzou, Katerina, Slavko Žitnik, Jorge Gracia, Dagmar Gromann, Maria Pia di Buono, Johanna Monti, and Maxim Ionov, editors, Proceedings of the 5th Conference on Language, Data and Knowledge: TermTrends 2025, pages 16–24, Naples, Italy, September. Unior Press. find
  27. Knowles, Rebecca, Samuel Larkin, and Chi-Kiu Lo. 2024. MSLC24: Further challenges for metrics on a wide landscape of translation quality. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Ninth Conference on Machine Translation, pages 475–491, Miami, Florida, USA, November. Association for Computational Linguistics. find
  28. Kocmi, Tom, Vilém Zouhar, Eleftherios Avramidis, Roman Grundkiewicz, Marzena Karpinska, Maja Popović, Mrinmaya Sachan, and Mariya Shmatova. 2024. Error span annotation: A balanced approach for human evaluation of machine translation. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Ninth Conference on Machine Translation, pages 1440–1453, Miami, Florida, USA, November. Association for Computational Linguistics. ↗ our card
  29. Kocmi, Tom, Ekaterina Artemova, Eleftherios Avramidis, Rachel Bawden, Ondřej Bojar, Konstantin Dranch, Anton Dvorkovich, Sergey Dukanov, Mark Fishel, Markus Freitag, Thamme Gowda, Roman Grundkiewicz, Barry Haddow, Marzena Karpinska, Philipp Koehn, Howard Lakougna, Jessica Lundin, Christof Monz, Kenton Murray, Masaaki Nagata, Stefano Perrella, Lorenzo Proietti, Martin Popel, Maja Popović, Parker Riley, Mariya Shmatova, Steinthór Steingrı́msson, Lisa Yankovskaya, and Vilém Zouhar. 2025. Findings of the WMT25 general machine translation shared task: Time to stop evaluating on easy test sets. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Tenth Conference on Machine Translation, pages 355–413, Suzhou, China, November. Association for Computational Linguistics. find our card
  30. Koehn, Philipp and Rebecca Knowles. 2017. Six challenges for neural machine translation. In Luong, Thang, Alexandra Birch, Graham Neubig, and Andrew Finch, editors, Proceedings of the First Workshop on Neural Machine Translation, pages 28–39, Vancouver, August. Association for Computational Linguistics. find
  31. Kojima, Takeshi, Shixiang (Shane) Gu, Machel Reid, Yutaka Matsuo, and Yusuke Iwasawa. 2022. Large Language Models are Zero-Shot Reasoners. In Koyejo, S., S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, editors, Advances in Neural Information Processing Systems, volume 35, pages 22199–22213. Curran Associates, Inc. find our card
  32. Lee, Jinu and Julia Hockenmaier. 2025. Evaluating step-by-step reasoning traces: A survey. In Christodoulopoulos, Christos, Tanmoy Chakraborty, Carolyn Rose, and Violet Peng, editors, Findings of the Association for Computational Linguistics: EMNLP 2025, pages 1789–1814, Suzhou, China, November. Association for Computational Linguistics. find
  33. Li, Yin and Zilong Zhong. 2024. Visual Insights into Translation: Demystifying Trends of Adopting Eye-Tracking Techniques in Translation Studies. Frontiers in Psychology, 15. ↗
  34. Li, Zihao, Shaoxiong Ji, and Jörg Tiedemann. 2026. Test-Time Scaling of Reasoning Models for Machine Translation. Arxiv:2510.06471. ↗
  35. Lindsey, Jack, Wes Gurnee, Emmanuel Ameisen, Brian Chen, Adam Pearce, Nicholas L. Turner, Craig Citro, David Abrahams, Shan Carter, Basil Hosmer, Jonathan Marcus, Michael Sklar, Adly Templeton, Trenton Bricken, Callum McDougall, Hoagy Cunningham, Thomas Henighan, Adam Jermyn, Andy Jones, Andrew Persic, Zhenyi Qi, T. Ben Thompson, Sam Zimmerman, Kelley Rivoire, Thomas Conerly, Chris Olah, and Joshua Batson. 2025. On the Biology of a Large Language Model. Transformer Circuits Thread. find
  36. Liu, Pengfei, Weizhe Yuan, Jinlan Fu, Zhengbao Jiang, Hiroaki Hayashi, and Graham Neubig. 2023. Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing. ACM Comput. Surv., 55(9), January. ↗ our card
  37. Liu, Sinuo, Chenyang Lyu, Minghao Wu, Longyue Wang, Weihua Luo, Kaifu Zhang, and Zifu Shang. 2025a. New Trends for Modern Machine Translation with Large Reasoning Models. find our card
  38. Liu, Zichen, Changyu Chen, Wenjun Li, Penghui Qi, Tianyu Pang, Chao Du, Wee Sun Lee, and Min Lin. 2025b. Understanding R1-Zero-Like Training: A Critical Perspective. In Second Conference on Language Modeling. ↗ our card
  39. Lyu, Chenyang, Zefeng Du, Jitao Xu, Yitao Duan, Minghao Wu, Teresa Lynn, Alham Fikri Aji, Derek F. Wong, and Longyue Wang. 2024. A paradigm shift: The future of machine translation lies with large language models. In Calzolari, Nicoletta, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, and Nianwen Xue, editors, Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 1339–1352, Torino, Italia, May. ELRA and ICCL. find
  40. Ma, Wenjie, Jingxuan He, Charlie Snell, Tyler Griggs, Sewon Min, and Matei Zaharia. 2025. Reasoning Models Can Be Effective Without Thinking. Arxiv:2504.09858. ↗
  41. Marjanović, Sara Vera, Arkil Patel, Vaibhav Adlakha, Milad Aghajohari, Parishad BehnamGhader, Mehar Bhatia, Aditi Khandelwal, Austin Kraft, Benno Krojer, Xing Han Lù, Nicholas Meade, Dongchan Shin, Amirhossein Kazemnejad, Gaurav Kamath, Marius Mosbach, Karolina Stańczak, and Siva Reddy. 2026. DeepSeek-R1 Thoughtology: Let’s think about LLM Reasoning. arXiv: 2504.07128. ↗
  42. Martins, Pedro Henrique, Patrick Fernandes, João Alves, Nuno M. Guerreiro, Ricardo Rei, Duarte M. Alves, José Pombal, Amin Farajian, Manuel Faysse, Mateusz Klimaszewski, Pierre Colombo, Barry Haddow, José G. C. de Souza, Alexandra Birch, and André F. T. Martins. 2024. EuroLLM: Multilingual language models for europe. Arxiv:2409.16235. ↗ our card
  43. Molfese, Francesco Maria, Luca Moroni, Ciro Porcaro, Simone Conia, and Roberto Navigli. 2026. Retraceqa: Evaluating reasoning traces of small language models in commonsense question answering. Arxiv:2510.09351. ↗
  44. Muennighoff, Niklas, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candès, and Tatsunori Hashimoto. 2025. s1: Simple test-time scaling. Arxiv:2501.19393. ↗ our card
  45. Nye, Maxwell, Anders Johan Andreassen, Guy Gur-Ari, Henryk Michalewski, Jacob Austin, David Bieber, David Dohan, Aitor Lewkowycz, Maarten Bosma, David Luan, Charles Sutton, and Augustus Odena. 2021. Show Your Work: Scratchpads for Intermediate Computation with Language Models. Arxiv:2112.00114. ↗ our card
  46. OpenAI. 2024. OpenAI o1 System Card. 2412.16720. ↗ our card
  47. Paglieri, Davide, Bartłomiej Cupiał, Samuel Coward, Ulyana Piterbarg, Maciej Wolczyk, Akbir Khan, Eduardo Pignatelli, Łukasz Kuciński, Lerrel Pinto, Rob Fergus, Jakob Nicolaus Foerster, Jack Parker-Holder, and Tim Rocktäschel. 2025. BALROG: Benchmarking Agentic LLM and vlm reasoning on games. Arxiv:2411.13543. ↗
  48. Pang, Jianhui, Fanghua Ye, Derek Fai Wong, Dian Yu, Shuming Shi, Zhaopeng Tu, and Longyue Wang. 2025. Salute the classic: Revisiting challenges of machine translation in the age of large language models. Transactions of the Association for Computational Linguistics, 13:73–95. ↗
  49. Peng, ShengYun, Eric Smith, Ivan Evtimov, Song Jiang, Pin-Yu Chen, Hongyuan Zhan, Haozhu Wang, Duen Horng Chau, Mahesh Pasupuleti, and Jianfeng Chi. 2025. Large Reasoning Models Learn Better Alignment from Flawed Thinking. Arxiv:2510.00938. ↗
  50. Perrella, Stefano, Lorenzo Proietti, Pere-Lluı́s Huguet Cabot, Edoardo Barba, and Roberto Navigli. 2024. Beyond correlation: Interpretable evaluation of machine translation metrics. In Al-Onaizan, Yaser, Mohit Bansal, and Yun-Nung Chen, editors, Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 20689–20714, Miami, Florida, USA, November. Association for Computational Linguistics. find our card
  51. Pombal, José, Nuno M. Guerreiro, Ricardo Rei, and André F. T. Martins. 2025. Adding Chocolate to Mint: Mitigating Metric Interference in Machine Translation. Transactions of the Association for Computational Linguistics, 13:1319–1339, 10. ↗
  52. Press, Ofir, Muru Zhang, Sewon Min, Ludwig Schmidt, Noah Smith, and Mike Lewis. 2023. Measuring and narrowing the compositionality gap in language models. In Bouamor, Houda, Juan Pino, and Kalika Bali, editors, Findings of the Association for Computational Linguistics: EMNLP 2023, pages 5687–5711, Singapore, December. Association for Computational Linguistics. find
  53. Proietti, Lorenzo, Stefano Perrella, Vilém Zouhar, Roberto Navigli, and Tom Kocmi. 2025. Estimating machine translation difficulty. In Christodoulopoulos, Christos, Tanmoy Chakraborty, Carolyn Rose, and Violet Peng, editors, Findings of the Association for Computational Linguistics: EMNLP 2025, pages 24261– 24285, Suzhou, China, November. Association for Computational Linguistics. ↗
  54. Qin, Yiwei, Xuefeng Li, Haoyang Zou, Yixiu Liu, Shijie Xia, Zhen Huang, Yixin Ye, Weizhe Yuan, Hector Liu, Yuanzhi Li, and Pengfei Liu. 2024. O1 Replication Journey: A Strategic Progress Report – Part 1. Arxiv:2410.18982. ↗
  55. Raganato, Alessandro, Yves Scherrer, and Jörg Tiedemann. 2020. An evaluation benchmark for testing the word sense disambiguation capabilities of machine translation systems. In Calzolari, Nicoletta, Frédéric Béchet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Hélène Mazo, Asuncion Moreno, Jan Odijk, and Stelios Piperidis, editors, Proceedings of the Twelfth Language Resources and Evaluation Conference, pages 3668–3675, Marseille, France, May. European Language Resources Association. find
  56. Ralli, Natascia and Norbert Andreatta. 2018. bistro – ein Tool für mehrsprachige Rechtsterminologie. trans-kom, 11(1):7–44. find
  57. Rei, Ricardo, Craig Stewart, Ana C Farinha, and Alon Lavie. 2020. COMET: A neural framework for MT evaluation. In Webber, Bonnie, Trevor Cohn, Yulan He, and Yang Liu, editors, Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2685–2702, Online, November. Association for Computational Linguistics. find our card
  58. Schaeffer, Moritz, Jean Nitzke, Anke Tardel, Katharina Oster, Silke Gutermuth, and Silvia Hansen-Schirra. 2019. Eye-tracking Revision Processes of Translation Students and Professional Translators. Perspectives, 27:589–603. ↗
  59. Semenov, Kirill and Ondřej Bojar. 2022. Automated evaluation metric for terminology consistency in MT. In Koehn, Philipp, Loı̈c Barrault, Ondřej Bojar, Fethi Bougares, Rajen Chatterjee, Marta R. Costa-jussà, Christian Federmann, Mark Fishel, Alexander Fraser, Markus Freitag, Yvette Graham, Roman Grundkiewicz, Paco Guzman, Barry Haddow, Matthias Huck, Antonio Jimeno Yepes, Tom Kocmi, André Martins, Makoto Morishita, Christof Monz, Masaaki Nagata, Toshiaki Nakazawa, Matteo Negri, Aurélie Névéol, Mariana Neves, Martin Popel, Marco Turchi, and Marcos Zampieri, editors, Proceedings of the Seventh Conference on Machine Translation (WMT), pages 450– 457, Abu Dhabi, United Arab Emirates (Hybrid), December. Association for Computational Linguistics. find
  60. Semenov, Kirill, Vilém Zouhar, Tom Kocmi, Dongdong Zhang, Wangchunshu Zhou, and Yuchen Eleanor Jiang. 2023. Findings of the wmt 2023 shared task on machine translation with terminologies. In Proceedings of the Eight Conference on Machine Translation (WMT). Association for Computational Linguistics, December. find
  61. Semenov, Kirill, Xu Huang, Vilém Zouhar, Nathaniel Berger, Dawei Zhu, Arturo Oncevay, and Pinzhen Chen. 2025. Findings of the WMT25 terminology translation task: Terminology is useful especially for good MTs. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Tenth Conference on Machine Translation, pages 554– 576, Suzhou, China, November. Association for Computational Linguistics. find
  62. Shah, Darsh J, Peter Rushton, Somanshu Singla, Mohit Parmar, Kurt Smith, Yash Vanjani, Ashish Vaswani, Adarsh Chaluvaraju, Andrew Hojel, Andrew Ma, Anil Thomas, Anthony Polloreno, Ashish Tanwer, Burhan Drak Sibai, Divya S Mansingka, Divya Shivaprasad, Ishaan Shah, Karl Stratos, Khoi Nguyen, Michael Callahan, Michael Pust, Mrinal Iyer, Philip Monk, Platon Mazarakis, Ritvik Kapila, Saurabh Srivastava, and Tim Romanski. 2025. Rethinking Reflection in Pre-Training. Arxiv:2504.04022. ↗
  63. Shrivastava, Vaishnavi, Ahmed Awadallah, Vidhisha Balachandran, Shivam Garg, Harkirat Behl, and Dimitris Papailiopoulos. 2025. Sample More to Think Less: Group Filtered Policy Optimization for Concise Reasoning. Arxiv:2508.09726. ↗
  64. Snell, Charlie Victor, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. 2024. Scaling LLM test-time compute optimally can be more effective than scaling model parameters. ArXiv, abs/2408.03314. ↗ our card
  65. Sun, Sanjun. 2011. Think-Aloud-Based Translation Process Research: Some Methodological Considerations. Meta, 56(4):928–951, December. ↗
  66. Sutton, Richard S. and Andrew G. Barto. 2018. Reinforcement Learning: An Introduction. A Bradford Book, Cambridge, MA, USA. find
  67. Tan, Shaomu, Ryosuke Mitani, Ritvik Choudhary, and Toshiyuki Sekiya. 2025. Investigating Test-Time Scaling with Reranking for Machine Translation. Arxiv:2509.19020. ↗
  68. Tan, Shaomu. 2025. Simple test time scaling for machine translation: Kaze-MT at the WMT25 general translation task. In Haddow, Barry, Tom Kocmi, Philipp Koehn, and Christof Monz, editors, Proceedings of the Tenth Conference on Machine Translation, pages 651–656, Suzhou, China, November. Association for Computational Linguistics. find
  69. Tran, Khanh-Tung, Barry O’Sullivan, and Hoang Nguyen. 2024. Irish-based large language model with extreme low-resource settings in machine translation. In Ojha, Atul Kr., Chao-hong Liu, Ekaterina Vylomova, Flammie Pirinen, Jade Abbott, Jonathan Washington, Nathaniel Oco, Valentin Malykh, Varvara Logacheva, and Xiaobing Zhao, editors, Proceedings of the Seventh Workshop on Technologies for Machine Translation of Low-Resource Languages (LoResMT 2024), pages 193–202, Bangkok, Thailand, August. Association for Computational Linguistics. find
  70. Tuggener, Don. 2016. Incremental Coreference Resolution for German. Master’s thesis, University of Zurich, Faculty of Arts. ↗
  71. Turpin, Miles, Julian Michael, Ethan Perez, and Samuel R. Bowman. 2023. Language models don’t always say what they think: unfaithful explanations in chain-of-thought prompting. In Proceedings of the 37th International Conference on Neural Information Processing Systems, NIPS ’23, Red Hook, NY, USA. Curran Associates Inc. find our card
  72. Wang, Jiaan, Fandong Meng, Yunlong Liang, and Jie Zhou. 2025a. DRT: Deep Reasoning Translation via Long Chain-of-Thought. Arxiv:2412.17498. ↗
  73. Wang, Jiaan, Fandong Meng, and Jie Zhou. 2025b. DeepTrans: Deep Reasoning Translation via Reinforcement Learning. Arxiv:2504.10187. ↗
  74. Wang, Jiaan, Fandong Meng, and Jie Zhou. 2025c. ExTrans: Multilingual Deep Reasoning Translation via Exemplar-Enhanced Reinforcement Learning. Arxiv:2505.12996. ↗
  75. Wang, Rui, Hongru Wang, Boyang Xue, Jianhui Pang, Shudong Liu, Yi Chen, Jiahao Qiu, Derek Fai Wong, Heng Ji, and Kam-Fai Wong. 2025d. Harnessing the Reasoning Economy: A Survey of Efficient Reasoning for Large Language Models. Arxiv:2503.24377. ↗
  76. Wang, Yifang, Saihong Li, and Yubo Zhou Rasmussen. 2025e. Translators’ allocation of cognitive resources in two translation directions: A study using eye-tracking and keystroke logging. Applied Sciences, 15(8). ↗
  77. Wang, Jun. 2025. A Tutorial on LLM reasoning: Relevant methods behind chatgpt o1. Arxiv:2502.10867. ↗
  78. Wassie, Aman Kassahun, Mahdi Molaei, and Yasmin Moslem. 2025. Domain-specific translation with open-source large language models: Resource-oriented analysis. ↗
  79. Wu, Zhanglin, Daimeng Wei, Xiaoyu Chen, Hengchao Shang, Jiaxin Guo, Zongyao Li, Yuanchang Luo, Jinlong Yang, Zhiqiang Rao, and Hao Yang. 2025. Combining the best of both worlds: A method for hybrid NMT and LLM translation. In Che, Wanxiang, Joyce Nabende, Ekaterina Shutova, and Mohammad Taher Pilehvar, editors, Findings of the Association for Computational Linguistics: ACL 2025, pages 5140–5148, Vienna, Austria, July. Association for Computational Linguistics. ↗
  80. Wurgaft, Daniel, Ben Prystawski, Kanishk Gandhi, Cedegao E. Zhang, Joshua B. Tenenbaum, and Noah D. Goodman. 2025. Scaling up the think-aloud method. Arxiv:2505.23931. ↗
  81. Xiao, Hongfei. 2025. Application of the Think-Aloud Protocols in Translation Studies: Experience, Problems and Prospects. ISPCE Bulletin. Jiangxi Normal University Science and Technology College, China. ↗
  82. Yao, Zijun, Yantao Liu, Yanxu Chen, Jianhui Chen, Junfeng Fang, Lei Hou, Juanzi Li, and Tat-Seng Chua. 2025. Are Reasoning Models More Prone to Hallucination? Arxiv:2505.23646. ↗
  83. Ye, Yongshi, Biao Fu, Chongxuan Huang, Yidong Chen, and Xiaodong Shi. 2025. How Well Do Large Reasoning Models Translate? A Comprehensive Evaluation for Multi-Domain Machine Translation. Arxiv:2505.19987. ↗
  84. Yeo, Edward, Yuxuan Tong, Morry Niu, Graham Neubig, and Xiang Yue. 2025. Demystifying Long Chain-of-Thought Reasoning in LLMs. Arxiv:2502.03373. ↗ our card
  85. Yue, Richard, John Ortega, and Kenneth Church. 2024. On translating technical terminology: A translation workflow for machine-translated acronyms. In Knowles, Rebecca, Akiko Eriguchi, and Shivali Goel, editors, Proceedings of the 16th Conference of the Association for Machine Translation in the Americas (Volume 1: Research Track), pages 48–54, Chicago, USA, September. Association for Machine Translation in the Americas. ↗
  86. Zaman, Kerem and Shashank Srivastava. 2025. Is Chain-of-Thought Really Not Explainability? Chain-of-thought can be faithful without hint verbalization. Arxiv:2512.23032. ↗
  87. Zampieri, Marcos, Kai North, Tommi Jauhiainen, Mariano Felice, Neha Kumari, Nishant Nair, and Yash Mahesh Bangera. 2024. Language variety identification with true labels. In Calzolari, Nicoletta, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, and Nianwen Xue, editors, Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 10100–10109, Torino, Italia, May. ELRA and ICCL. find
  88. Zhang, Huaao, Qiang Wang, Bo Qin, Zelin Shi, Haibo Wang, and Ming Chen. 2023. Understanding and improving the robustness of terminology constraints in neural machine translation. In Rogers, Anna, Jordan Boyd-Graber, and Naoaki Okazaki, editors, Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 6029–6042, Toronto, Canada, July. Association for Computational Linguistics. find
  89. Zhang, Jiajie, Nianyi Lin, Lei Hou, Ling Feng, and Juanzi Li. 2025. AdaptThink: Reasoning models can learn when to think. In Christodoulopoulos, Christos, Tanmoy Chakraborty, Carolyn Rose, and Violet Peng, editors, Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 3716–3730, Suzhou, China, November. Association for Computational Linguistics. find
  90. Zhao, Chengshuai, Zhen Tan, Pingchuan Ma, Dawei Li, Bohan Jiang, Yancheng Wang, Yingzhen Yang, and Huan Liu. 2026. Is Chain-of-Thought Reasoning of LLMs a Mirage? A Data Distribution Lens. Arxiv:2508.01191. ↗

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/natale-2026-beyond-simple-term-injection-reasoning/