Elamur Research⌕
☰
home / papers / 2026 / NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning

NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning

Paper · 2026 · Zhongtao Miao, Kaiyan Zhao, Masaaki Nagata et al.

Summary

LLM-based translators struggle with sentences containing newly coined words, because the knowledge stored in their parameters is fixed at training time. Prompting and reasoning alone cannot fill that gap. The authors address this by building NeoAMT, an agent that alternates between reasoning and looking words up in a dictionary.

They mined an English Wiktionary dump of roughly 10M records to create two resources:

  • Neko, a neologism translation dataset covering 16 languages and 75 directions. Its test set uses human-verified Wiktionary examples plus reference-free English-to-other evaluation.
  • A search toolkit, a dense-retrieval dictionary built from about 3.3M cleaned entries.

An agent built on Qwen3-8B is trained with GRPO-based reinforcement learning. The training adds a new reward design and adaptive rollout sampling guided by a "translation difficulty" metric.

The resulting NeoAMT-8B scores notably higher than Qwen3-8B on XCOMET-XL across the Neko test directions, even when both models use the same toolkit. It also consults the dictionary more actively. The work offers a public, multilingual testbed for a largely neglected MT problem, including Ukrainian.

Key points

  • Neko size: 11,521 examples (10,425 train / 83 val / 1,013 test). The test set has 743 other-to-English pairs and 270 English-to-other pairs, the latter judged reference-free using neologism definitions. The earlier Neo-bench had only 240 non-public examples in one language pair.
  • Data pipeline: cleaning produced 3,312,877 records, 3,606 of them tagged as neologisms. These split into 831 entries with translated examples (Type 1), 2,581 entries with untranslated examples (Type 2) and the rest (Type 3). Training data comes from 700 Type 2 sentences that gpt-5-2025-08-07 translated into the other 15 languages, with aligned neologism spans.
  • Search toolkit: it embeds the full entry (part-of-speech, etymology, senses, translations) with bge-m3 and retrieves via faiss using cosine similarity. Ukrainian contributes 52,254 definitions, and uk-en is one of the test directions.
  • Behaviour analysis: on 100 zh-en samples, RL training increased use of the search tool. The most common reasoning pattern was "confirmation", where the model guesses a meaning and then verifies it by search.
  • Stated limitations: the general-purpose bge-m3 retriever leaves room for improvement, and the model's thinking does not always follow the instructed format, since the reward only constrains the final output.

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

 Institutions

🎓The University of TokyoTokyo · Japan96 works · 9,600 citations🏢NTTTokyo · Japan28 works · 5,199 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 18,614 worksArtificial Intelligence · 17,965 worksArts and Humanities · 17,450 worksLanguage and Linguistics · 17,296 works

🧭 Topics 1Research subjects and languages the work is about.Machine translation · 1,105 works

Cite

Cite as Zhongtao Miao, Kaiyan Zhao, Masaaki Nagata and Yoshimasa Tsuruoka. 2026. NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.acl-long.345

@inproceedings{miao2026neoamt,
  title = {{NeoAMT: Neologism-Aware Agentic Machine Translation with Reinforcement Learning}},
  author = {Zhongtao Miao, Kaiyan Zhao, Masaaki Nagata and Yoshimasa Tsuruoka},
  booktitle = {Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2026},
  doi = {10.18653/v1/2026.acl-long.345},
  url = {https://doi.org/10.18653/v1/2026.acl-long.345}
}

References

58 references, 30 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sandhini Agarwal, Lama Ahmad, Jason Ai, Sam Altman, Andy Applebaum, Edwin Arbus, Rahul K Arora, Yu Bai, Bowen Baker, Haiming Bao, and 1 others. 2025. gpt-oss-120b & gpt-oss-20b model card. arXiv preprint arXiv:2508.10925.
  2. Duarte Miguel Alves, José Pombal, Nuno M Guerreiro, Pedro Henrique Martins, João Alves, Amin Farajian, Ben Peters, Ricardo Rei, Patrick Fernandes, Sweta Agrawal, Pierre Colombo, José G. C. de Souza, and Andre Martins. 2024. Tower: An open multilingual large language model for translation-related tas…
  3. Jason Ansel, Edward Yang, Horace He, Natalia Gimelshein, Animesh Jain, Michael Voznesensky, Bin Bao, Peter Bell, David Berard, Evgeni Burovski, Geeta Chauhan, Anjali Chourdia, Will Constable, Alban Desmaison, Zachary DeVito, Elias Ellison, Will Feng, Jiong Gong, Michael Gschwind, and 30 others. 2024… — PyTorch 2: Faster Machine Learning Through Dynamic Python Bytecode Transformation and Graph Compilation
  4. Viraat Aryabumi, John Dang, Dwarak Talupuru, Saurabh Dash, David Cairuz, Hangyu Lin, Bharat Venkitesh, Madeline Smith, Jon Ander Campos, Yi Chern Tan, and 1 others. 2024. Aya 23: Open weight releases to further multilingual progress. arXiv preprint arXiv:2405.15032.
  5. Andong Chen, Lianzhang Lou, Kehai Chen, Xuefeng Bai, Yang Xiang, Muyun Yang, Tiejun Zhao, and Min Zhang. 2025. Benchmarking LLMs for translating classical Chinese poetry: Evaluating adequacy, fluency, and elegance. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Proces…
  6. Jianlyu Chen, Shitao Xiao, Peitian Zhang, Kun Luo, Defu Lian, and Zheng Liu. 2024. M3-embedding: Multi-linguality, multi-functionality, multi-granularity text embeddings through self-knowledge distillation. In Findings of the Association for Computational Linguistics: ACL 2024, pages 2318–2335, Bang…
  7. John Dang, Shivalika Singh, Daniel D’souza, Arash Ahmadian, Alejandro Salamanca, Madeline Smith, Aidan Peppin, Sungjin Hong, Manoj Govindassamy, Terrence Zhao, and 1 others. 2024. Aya expanse: Combining research breakthroughs for a new multilingual frontier. arXiv preprint arXiv:2412.04261.
  8. Daniel Deutsch, Eleftheria Briakou, Isaac Rayburn Caswell, Mara Finkelstein, Rebecca Galor, Juraj Juraska, Geza Kovacs, Alison Lui, Ricardo Rei, Jason Riesa, Shruti Rijhwani, Parker Riley, Elizabeth Salesky, Firas Trabelsi, Stephanie Winkler, Biao Zhang, and Markus Freitag. 2025. WMT24++: Expanding…
  9. Guanting Dong, Hangyu Mao, Kai Ma, Licheng Bao, Yifei Chen, Zhongyuan Wang, Zhongxia Chen, Jiazhen Du, Huiyang Wang, Fuzheng Zhang, and 1 others. 2025. Agentic reinforced policy optimization. arXiv preprint arXiv:2507.19849. — [arXiv:2507.19849](https://arxiv.org/abs/2507.19849) find
  10. Matthijs Douze, Alexandr Guzhva, Chengqi Deng, Jeff Johnson, Gergely Szilvasy, Pierre-Emmanuel Mazaré, Maria Lomeli, Lucas Hosseini, and Hervé Jégou. 2024. The faiss library.
  11. Zhaopeng Feng, Shaosheng Cao, Jiahan Ren, Jiayuan Su, Ruizhe Chen, Yan Zhang, Jian Wu, and Zuozhu Liu. 2025. MT-r1-zero: Advancing LLM-based machine translation via r1-zero-like reinforcement learning. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 18685– 18702, Suzh… find
  12. Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, and 1 others. 2024. The llama 3 herd of models. arXiv preprint arXiv:2407.21783.
  13. Nuno M. Guerreiro, Ricardo Rei, Daan van Stigt, Luisa Coheur, Pierre Colombo, and André F. T. Martins. 2024. xCOMET: Transparent machine translation evaluation through fine-grained error detection. Transactions of the Association for Computational Linguistics, 12:979–995.
  14. Daya Guo, 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, and 175 others. 2025. DeepSeek-R1 incentivizes reasoning in LLMs through reinfo… find
  15. Masoud Jalili Sabet, Philipp Dufter, François Yvon, and Hinrich Schütze. 2020. SimAlign: High quality word alignments without parallel training data using static and contextualized embeddings. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1627–1643, Online. Associat… find
  16. Bowen Jin, Hansi Zeng, Zhenrui Yue, Jinsung Yoon, Sercan O Arik, Dong Wang, Hamed Zamani, and Jiawei Han. 2025. Search-r1: Training LLMs to reason and leverage search engines with reinforcement learning. In Second Conference on Language Modeling.
  17. Jeff Johnson, Matthijs Douze, and Hervé Jégou. 2019. Billion-scale similarity search with GPUs. IEEE Transactions on Big Data, 7(3):535–547.
  18. Tom Kocmi and Christian Federmann. 2023. Large language models are state-of-the-art evaluators of translation quality. In Proceedings of the 24th Annual Conference of the European Association for Machine Translation, pages 193–203, Tampere, Finland. European Association for Machine Translation. find
  19. Woosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng, Lianmin Zheng, Cody Hao Yu, Joseph Gonzalez, Hao Zhang, and Ion Stoica. 2023. Efficient memory management for large language model serving with pagedattention. In Proceedings of the 29th Symposium on Operating Systems Principles, SOSP ’23, page 611…
  20. Gaetan Latouche, Marc-André Carbonneau, and Benjamin Swanson. 2024. BinaryAlign: Word alignment as binary sequence labeling. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 10277–10288, Bangkok, Thailand. Association for Compu… find
  21. Paul Lerner and François Yvon. 2025. Towards the machine translation of scientific neologisms. In Proceedings of the 31st International Conference on Computational Linguistics, pages 947–963, Abu Dhabi, UAE. Association for Computational Linguistics. find
  22. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020. Retrieval-augmented generation for knowledge-intensive nlp tasks. In Advances in Neural Information Proc…
  23. Xiaoxi Li, Guanting Dong, Jiajie Jin, Yuyao Zhang, Yujia Zhou, Yutao Zhu, Peitian Zhang, and Zhicheng Dou. 2025a. Search-o1: Agentic search-enhanced large reasoning models. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 5420–5438, Suzhou, China. Asso…
  24. Zheng Li, Mao Zheng, Mingyang Song, and Wenjie Yang. 2025b. Tat-r1: Terminology-aware translation with reinforcement learning and word alignment. arXiv preprint arXiv:2505.21172. — [arXiv:2505.21172](https://arxiv.org/abs/2505.21172) find
  25. Zihao Li, Shaoxiong Ji, and Jörg Tiedemann. 2026. Test-time scaling of reasoning models for machine translation. In Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2902–2917, Rabat, Morocco. Association for Co…
  26. Aixin Liu, Bei Feng, Bing Xue, Bingxuan Wang, Bochao Wu, Chengda Lu, Chenggang Zhao, Chengqi Deng, Chenyu Zhang, Chong Ruan, and 1 others. 2024. Deepseek-v3 technical report. arXiv preprint arXiv:2412.19437. find
  27. Zhongtao Miao, Qiyu Wu, Masaaki Nagata, and Yoshimasa Tsuruoka. 2025. Improving word alignment using semi-supervised learning. In Findings of the Association for Computational Linguistics: ACL 2025, pages 19871–19888, Vienna, Austria. Association for Computational Linguistics. find
  28. Zhongtao Miao, Qiyu Wu, Kaiyan Zhao, Zilong Wu, and Yoshimasa Tsuruoka. 2024. Enhancing cross-lingual sentence embedding for low-resource languages with word alignment. In Findings of the Association for Computational Linguistics: NAACL 2024, pages 3225–3236, Mexico City, Mexico. Association for Com… find
  29. Niklas Muennighoff, Zitong Yang, Weijia Shi, Xiang Lisa Li, Li Fei-Fei, Hannaneh Hajishirzi, Luke Zettlemoyer, Percy Liang, Emmanuel Candes, and Tatsunori Hashimoto. 2025. s1: Simple test-time scaling. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 2…
  30. Masaaki Nagata, Katsuki Chousa, and Masaaki Nishino. 2020. A supervised word alignment method based on cross-language span prediction using multilingual BERT. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 555–565, Online. Association for Com… find
  31. Martin A. Nowak and David C. Krakauer. 1999. The evolution of language. Proceedings of the National Academy of Sciences, 96(14):8028–8033. find
  32. OpenAI. 2024. Learning to reason with llms. find
  33. Ricardo Rei, Nuno M. Guerreiro, José Pombal, Daan van Stigt, Marcos Treviso, Luisa Coheur, José G. C. de Souza, and André Martins. 2023. Scaling up CometKiwi: Unbabel-IST 2023 submission for the quality estimation shared task. In Proceedings of the Eighth Conference on Machine Translation, pages 84…
  34. Holger Schwenk, Guillaume Wenzek, Sergey Edunov, Edouard Grave, Armand Joulin, and Angela Fan. 2021. CCMatrix: Mining billions of high-quality parallel sentences on the web. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Co…
  35. Kirill Semenov, 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 Proceedings of the Tenth Conference on Machine Translation, pages 554–576, Suzhou, China.… find
  36. Kirill Semenov, 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 Eighth Conference on Machine Translation, pages 663–671, Singapore. Association for Computa… find
  37. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, YK Li, Yang Wu, and 1 others. 2024. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. arXiv preprint arXiv:2402.03300.
  38. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2025. Hybridflow: A flexible and efficient rlhf framework. In Proceedings of the Twentieth European Conference on Computer Systems, EuroSys ’25, page 1279–1297, New York, NY, USA. Assoc…
  39. Charlie Victor Snell, Jaehoon Lee, Kelvin Xu, and Aviral Kumar. 2025. Scaling LLM test-time compute optimally can be more effective than scaling parameters for reasoning. In The Thirteenth International Conference on Learning Representations. find
  40. Saba Sturua, Isabelle Mohr, Mohammad Kalim Akram, Michael Günther, Bo Wang, Markus Krimmel, Feng Wang, Georgios Mastrapas, Andreas Koukounas, Nan Wang, and 1 others. 2024. jina-embeddings-v3: Multilingual embeddings with task lora. arXiv preprint arXiv:2409.10173.
  41. Kimi Team, Angang Du, Bofei Gao, Bowei Xing, Changjiu Jiang, Cheng Chen, Cheng Li, Chenjun Xiao, Chenzhuang Du, Chonghua Liao, and 1 others. 2025. Kimi k1. 5: Scaling reinforcement learning with llms. arXiv preprint arXiv:2501.12599. — Kimi k1.5: Scaling Reinforcement Learning with LLMs
  42. 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, and 20 o… find
  43. Jiaan Wang, Fandong Meng, Yunlong Liang, and Jie Zhou. 2025a. DRT: Deep reasoning translation via long chain-of-thought. In Findings of the Association for Computational Linguistics: ACL 2025, pages 6770–6782, Vienna, Austria. Association for Computational Linguistics. find
  44. Jiaan Wang, Fandong Meng, and Jie Zhou. 2025b. Deep reasoning translation via reinforcement learning. arXiv preprint arXiv:2504.10187. — [arXiv:2504.10187](https://arxiv.org/abs/2504.10187) find
  45. Jiaan Wang, Fandong Meng, and Jie Zhou. 2025c. Extrans: Multilingual deep reasoning translation via exemplar-enhanced reinforcement learning. arXiv preprint arXiv:2505.12996. — [arXiv:2505.12996](https://arxiv.org/abs/2505.12996) find
  46. Qiyu Wu, Masaaki Nagata, and Yoshimasa Tsuruoka. 2023. WSPAlign: Word alignment pre-training via large-scale weakly supervised span prediction. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 11084–11099, Toronto, Canada. Asso… find
  47. Haoran Xu, Young Jin Kim, Amr Sharaf, and Hany Hassan Awadalla. 2024a. A paradigm shift in machine translation: Boosting translation performance of large language models. In The Twelfth International Conference on Learning Representations. find
  48. Haoran Xu, Kenton Murray, Philipp Koehn, Hieu Hoang, Akiko Eriguchi, and Huda Khayrallah. 2025. X-ALMA: Plug & play modules and adaptive rejection for quality translation at scale. In The Thirteenth International Conference on Learning Representations. find
  49. Haoran Xu, Amr Sharaf, Yunmo Chen, Weiting Tan, Lingfeng Shen, Benjamin Van Durme, Kenton Murray, and Young Jin Kim. 2024b. Contrastive preference optimization: Pushing the boundaries of LLM performance in machine translation. In Forty-first International Conference on Machine Learning.
  50. An Yang, Anfeng Li, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chang Gao, Chengen Huang, Chenxu Lv, and 1 others. 2025a. Qwen3 technical report. arXiv preprint arXiv:2505.09388.
  51. Wenjie Yang, Mao Zheng, Mingyang Song, Zheng Li, and Sitong Wang. 2025b. Ssr-zero: Simple self-rewarding reinforcement learning for machine translation. arXiv preprint arXiv:2505.16637.
  52. Binwei Yao, Ming Jiang, Tara Bobinac, Diyi Yang, and Junjie Hu. 2024. Benchmarking machine translation with cultural awareness. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 13078–13096, Miami, Florida, USA. Association for Computational Linguistics.
  53. Tatu Ylonen. 2022. Wiktextract: Wiktionary as machine-readable structured data. In Proceedings of the Thirteenth Language Resources and Evaluation Conference, pages 1317–1325, Marseille, France. European Language Resources Association. find
  54. Yanzhao Zhang, Mingxin Li, Dingkun Long, Xin Zhang, Huan Lin, Baosong Yang, Pengjun Xie, An Yang, Dayiheng Liu, Junyang Lin, and 1 others. 2025. Qwen3 embedding: Advancing text embedding and reranking through foundation models. arXiv preprint arXiv:2506.05176.
  55. Kaiyan Zhao, Qiyu Wu, Xin-Qiang Cai, and Yoshimasa Tsuruoka. 2024. Leveraging multi-lingual positive instances in contrastive learning to improve sentence embedding. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)… find
  56. Kaiyan Zhao, Zheyong Xie, Zhongtao Miao, Xinze Lyu, Yao Hu, and Shaosheng Cao. 2026. Benchmarking machine translation on chinese social media texts. Preprint, arXiv:2601.22931. find
  57. Jonathan Zheng, Alan Ritter, and Wei Xu. 2024. NEO-BENCH: Evaluating robustness of large language models with neologisms. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 13885– 13906, Bangkok, Thailand. Association for Computa… find
  58. Mao Zheng, Zheng Li, Bingxin Qu, Mingyang Song, Yang Du, Mingrui Sun, and Di Wang. 2025. Hunyuan-mt technical report. arXiv preprint arXiv:2509.05209.

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/miao-2026-neoamt-neologism-aware-agentic-machine/