Elamur Research⌕
☰
home / papers / 2026 / Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning

Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning

Paper · 2026 · Lintang Sutawika, Gokul Swamy, Steven Wu et al.

Summary

Reasoning LLMs trained mostly on English chains of thought perform much worse when the same questions are posed in less-represented languages. Fixing this is difficult because target-language data is scarce and sparse answer-level rewards give RL little to learn from. The authors propose SP3F (Self-Play with Privileged Pairwise Feedback), which needs only English question–reference pairs.

SP3F has two stages:

  1. SFT stage: Qwen2.5-7B is fine-tuned on machine-translated versions of the English pairs.
  2. RL stage (GRPO): Verifiable rewards are combined with feedback from an LLM judge. The judge sees the English reference and decides which of two candidate responses matches it more closely. Each response's reward is its average win rate against the other samples, a self-play design that tolerates the judge's intransitive preferences.

Trained on data covering 18 languages, the resulting SP3F-7B beats Qwen2.5-7B-Instruct in both accuracy and language fidelity. It does so on in-domain math tasks and on out-of-domain non-math tasks, using roughly one eighth of the post-training data. It also generalizes better to unseen languages. The work shows a data-efficient route to multilingual reasoning that reuses English supervision.

Key points

  • Data efficiency: SP3F-7B used about 125K training samples, compared with about 1M for Qwen2.5-7B-Instruct.
  • Benchmarks: Evaluation covers MGSM and MT Math100 (math) plus Belebele and Global MMLU Lite (non-math). Both accuracy and language fidelity are reported.
  • Reward design: The reward sums four terms:
    • binary answer correctness;
    • binary \boxed{} formatting;
    • binary language fidelity, set to 1 when at least 70% of the response is classified as the target language (a binary signal discourages reward hacking with very short replies);
    • the pairwise win rate from the privileged judge.
  • Effect of privileged information: Giving the judge the English reference makes its preferences less intransitive. It also helps the judge spot sound reasoning chains even when the final answer is wrong. A judge without this privilege still prefers models trained with the privileged judge over models it trained itself.
  • Resources and gaps: Code is at github.com/lintangsutawika/sp3f, with artifacts in the Hugging Face collection neulab/sp3f. Improvements are smallest for some low-resource languages, such as Swahili and Telugu.

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

🎓Carnegie Mellon UniversityPittsburgh · United States539 works · 179,533 citations

 Fields, topics and concepts

📚 Shelves 4The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 works

🧭 Topics 2Research subjects and languages the work is about.Reasoning · 1,426 worksMultilingual & cross-lingual NLP · 828 works

Cite

Cite as Lintang Sutawika, Gokul Swamy, Steven Wu and Graham Neubig. 2026. Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning. 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.393

@inproceedings{sutawika2026gained,
  title = {{Gained in Translation: Privileged Pairwise Judges Enhance Multilingual Reasoning}},
  author = {Lintang Sutawika, Gokul Swamy, Steven Wu and Graham Neubig},
  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.393},
  url = {https://doi.org/10.18653/v1/2026.acl-long.393}
}

References

62 references, 27 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, and Sara Hooker. 2024. Back to basics: Revisiting reinforce style optimization for learning from human feedback in llms. arXiv preprint arXiv:2402.14740. find
  2. Mikel Artetxe, Vedanuj Goswami, Shruti Bhosale, Angela Fan, and Luke Zettlemoyer. 2023. Revisiting machine translation for cross-lingual classification. Preprint, arXiv:2305.14240. find
  3. Mohammad Gheshlaghi Azar, Zhaohan Daniel Guo, Bilal Piot, Remi Munos, Mark Rowland, Michal Valko, and Daniele Calandriello. 2024. A general theoretical paradigm to understand learning from human preferences. In International Conference on Artificial Intelligence and Statistics, pages 4447–4455. PMLR…
  4. Lucas Bandarkar, Davis Liang, Benjamin Muller, Mikel Artetxe, Satya Narayan Shukla, Donald Husa, Naman Goyal, Abhinandan Krishnan, Luke Zettlemoyer, and Madian Khabsa. 2024. The belebele benchmark: a Josh Barua, Seun Eisape, Kayo Yin, and Alane Suhr. 2025. Long chain-of-thought reasoning across lang… — [arXiv:2508.14828](https://arxiv.org/abs/2508.14828) find
  5. BigScience Workshop, Teven Le Scao, Angela Fan, Christopher Akiki, Ellie Pavlick, Suzana Ilić, Daniel Hesslow, Roman Castagné, Alexandra Sasha Luccioni, François Yvon, Matthias Gallé, Jonathan Tow, Alexander M. Rush, Stella Biderman, Albert Webson, Pawan Sasanka Ammanamanchi, Thomas Wang, Benoît Sa… find
  6. Samuel Cahyawijaya, Holy Lovenia, Alham Fikri Aji, Genta Winata, Bryan Wilie, Fajri Koto, Rahmad Mahendra, Christian Wibisono, Ade Romadhony, Karissa Vincentio, Jennifer Santoso, David Moeljadi, Cahya Wirawan, Frederikus Hudi, Muhammad Satrio Wicaksono, Ivan Parmonangan, Ika Alfina, Ilham Firdausi P… find
  7. Samuel Cahyawijaya, Holy Lovenia, and Pascale Fung. 2024. Llms are few-shot in-context low-resource language learners. In North American Chapter of the Association for Computational Linguistics.
  8. Dian Chen, Brady Zhou, Vladlen Koltun, and Philipp Krähenbühl. 2019. Learning by cheating. Preprint, arXiv:1912.12294. — [arXiv:1912.12294](https://arxiv.org/abs/1912.12294) find
  9. Sanjiban Choudhury, Mohak Bhardwaj, Sankalp Arora, Ashish Kapoor, Gireeja Ranade, Sebastian Scherer, and Debadeepta Dey. 2017. Data-driven planning via imitation learning. Preprint, arXiv:1711.06391. — [arXiv:1711.06391](https://arxiv.org/abs/1711.06391) find
  10. DeepSeek-AI, Daya Guo, Dejian Yang, Haowei Zhang, Junxiao Song, Ruoyu Zhang, Runxin Xu, Qihao Zhu, Shirong Ma, Peiyi Wang, Xiao Bi, Xiaokang Zhang, Xingkai Yu, Yu Wu, Z. F. Wu, Zhibin Gou, Zhihong Shao, Zhuoshu Li, Ziyi Gao, and 181 others. 2025. Deepseek-r1: Incentivizing reasoning capability in ll…
  11. Julen Etxaniz, Gorka Azkune, Aitor Soroa, Oier Lopez de Lacalle, and Mikel Artetxe. 2023. Do multilingual language models think better in english? Preprint, arXiv:2308.01223.
  12. Zhaolin Gao, Jonathan Chang, Wenhao Zhan, Owen Oertell, Gokul Swamy, Kianté Brantley, Thorsten Joachims, Drew Bagnell, Jason D Lee, and Wen Sun. 2024a. Rebel: Reinforcement learning via regressing relative rewards. Advances in Neural Information Processing Systems, 37:52354–52400. find
  13. Zhaolin Gao, Wenhao Zhan, Jonathan D Chang, Gokul Swamy, Kianté Brantley, Jason D Lee, and Wen Sun. 2024b. Regressing the relative future: Efficient policy optimization for multi-turn rlhf. arXiv preprint arXiv:2410.04612. — [arXiv:2410.04612](https://arxiv.org/abs/2410.04612) find
  14. Akash Ghosh, Debayan Datta, Sriparna Saha, and Chirag Agarwal. 2025. A survey of multilingual reasoning in language models. Preprint, arXiv:2502.09457. find
  15. Dylan Hadfield-Menell, Smitha Milli, Pieter Abbeel, Stuart J Russell, and Anca Dragan. 2017. Inverse reward design. Advances in neural information processing systems, 30. find
  16. Shulin Huang, Yiran Ding, Junshu Pan, and Yue Zhang. 2025. Beyond english-centric training: How reinforcement learning improves cross-lingual reasoning in llms. Preprint, arXiv:2509.23657. find
  17. Zixian Huang, Wenhao Zhu, Gong Cheng, Lei Li, and Fei Yuan. 2024. Mindmerger: Efficient boosting llm reasoning in non-english languages. ArXiv, abs/2405.17386. — [arXiv:2405.17386](https://arxiv.org/abs/2405.17386) find
  18. Jaedong Hwang, Kumar Tanmay, Seok-Jin Lee, Ayush Agrawal, Hamid Palangi, Kumar Ayush, I. Fiete, and Paul Pu Liang. 2025. Learn globally, speak locally: Bridging the gaps in multilingual reasoning. ArXiv, abs/2507.05418. — [arXiv:2507.05418](https://arxiv.org/abs/2507.05418) find
  19. Aaron Jaech, Adam Kalai, Adam Lerer, Adam Richardson, Ahmed El-Kishky, Aiden Low, Alec Helyar, Aleksander Madry, Alex Beutel, Alex Carney, and 1 others. 2024. Openai o1 system card. arXiv preprint arXiv:2412.16720.
  20. Arnav Kumar Jain, Vibhakar Mohta, Subin Kim, Atiksh Bhardwaj, Juntao Ren, Yunhai Feng, Sanjiban Choudhury, and Gokul Swamy. 2025. A smooth sea never made a skilled sailor: Robust imitation via learning to search. In The Thirty-ninth Annual Conference on Neural Information Processing Systems. find
  21. Pratik Joshi, Sebastin Santy, Amar Budhiraja, Kalika Bali, and Monojit Choudhury. 2020. The state and fate of linguistic diversity and inclusion in the NLP world. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 6282–6293, Online. Association for Comp…
  22. Sham Machandranath Kakade. 2003. On the sample complexity of reinforcement learning. University of London, University College London (United Kingdom). find
  23. Muhammad Dehan Al Kautsar, Aswin Candra, Muhammad Alif Al Hakim, Maxalmina Satria Kahfi, Fajri Koto, Alham Fikri Aji, Peerat Limkonchotiwat, Ekapol Chuangsuwanich, and Genta Indra Winata. 2025. Seadialogues: A multilingual culturally grounded multi-turn dialogue dataset on southeast asian languages.… — [arXiv:2508.07069](https://arxiv.org/abs/2508.07069) find
  24. Seungone Kim, Jamin Shin, Yejin Cho, Joel Jang, Shayne Longpre, Hwaran Lee, Sangdoo Yun, Seongjin Shin, Sungdong Kim, James Thorne, and Minjoon Seo. 2024. Prometheus: Inducing fine-grained evaluation capability in language models. Preprint, arXiv:2310.08491.
  25. Ashish Kumar, Zipeng Fu, Deepak Pathak, and Jitendra Malik. 2021. Rma: Rapid motor adaptation for legged robots. Preprint, arXiv:2107.04034. find
  26. Huiyuan Lai and Malvina Nissim. 2024. mCoT: Multilingual instruction tuning for reasoning consistency in language models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12012– 12026, Bangkok, Thailand. Association for Computa…
  27. Nathan Lambert, Jacob Morrison, Valentina Pyatkin, Shengyi Huang, Hamish Ivison, Faeze Brahman, Lester James V. Miranda, Alisa Liu, Nouha Dziri, Shane Lyu, Yuling Gu, Saumya Malik, Victoria Graf, Jena D. Hwang, Jiangjiang Yang, Ronan Le Bras, Oyvind Tafjord, Chris Wilhelm, Luca Soldaini, and 4 other… — Tulu 3: Pushing Frontiers in Open Language Model Post-Training
  28. Hunter Lightman, Vineet Kosaraju, Yura Burda, Harri Edwards, Bowen Baker, Teddy Lee, Jan Leike, John Schulman, Ilya Sutskever, and Karl Cobbe. 2023. Let’s verify step by step. Preprint, arXiv:2305.20050.
  29. Mingjie Liu, Shizhe Diao, Ximing Lu, Jian Hu, Xin Dong, Yejin Choi, Jan Kautz, and Yi Dong. 2025a. Prorl: Prolonged reinforcement learning expands reasoning boundaries in large language models. Preprint, arXiv:2505.24864. find
  30. Zichen Liu, 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. Preprint, arXiv:2503.20783.
  31. Michael Luo, Sijun Tan, Justin Wong, Xiaoxiang Shi, William Tang, Manan Roongta, Colin Cai, Jeffrey Luo, Tianjun Zhang, Erran Li, Raluca Ada Popa, and Ion Stoica. 2025. Deepscaler: Surpassing o1-preview with a 1.5b model by scaling rl. Notion Blog. find
  32. Niklas Muennighoff, Thomas Wang, Lintang Sutawika, Adam Roberts, Stella Biderman, Teven Le Scao, M Saiful Bari, Sheng Shen, Zheng-Xin Yong, Hailey Schoelkopf, Xiangru Tang, Dragomir Radev, Alham Fikri Aji, Khalid Almubarak, Samuel Albanie, Zaid Alyafeai, Albert Webson, Edward Raff, and Colin Raffel.… — Crosslingual Generalization through Multitask Finetuning
  33. Raymond Ng, Thanh Ngan Nguyen, Yuli Huang, Ngee Chia Tai, Wai Yi Leong, Wei Qi Leong, Xianbin Yong, Jian Gang Ngui, Yosephine Susanto, Nicholas Cheng, Hamsawardhini Rengarajan, Peerat Limkonchotiwat, Adithya Venkatadri Hulagadri, Kok Wai Teng, Yeo Yeow Tong, Bryan Siow, Wei Yi Teo, Wayne Lau, Choon… — SEA-LION: Southeast Asian Languages in One Network
  34. O Oertell, W Zhan, G Swamy, ZS Wu, K Brantley, J Lee, and W Sun. 2025. Heuristics considered harmful: Rl with random rewards should not make llms reason. Notion Blog. find
  35. Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Christiano, Jan Leike, and Ryan Lowe. 2022. Train…
  36. Arjun Panickssery, Samuel Bowman, and Shi Feng. 2024. Llm evaluators recognize and favor their own generations. Advances in Neural Information Processing Systems, 37:68772–68802. find
  37. Edoardo Maria Ponti, Julia Kreutzer, Ivan Vulić, and Siva Reddy. 2021. Modelling latent translations for cross-lingual transfer. Preprint, arXiv:2107.11353. — [arXiv:2107.11353](https://arxiv.org/abs/2107.11353) find
  38. Zhen Qin, Rolf Jagerman, Kai Hui, Honglei Zhuang, Junru Wu, Le Yan, Jiaming Shen, Tianqi Liu, Jialu Liu, Donald Metzler, and 1 others. 2024. Large language models are effective text rankers with pairwise ranking prompting. In Findings of the Association for Computational Linguistics: NAACL 2024, pag…
  39. Qwen, An Yang, Baosong Yang, Beichen Zhang, Binyuan Hui, Bo Zheng, Bowen Yu, Chengyuan Li, Dayiheng Liu, Fei Huang, Haoran Wei, Huan Lin, Jian Yang, Jianhong Tu, Jianwei Zhang, Jianxin Yang, Jiaxi Yang, Jingren Zhou, Junyang Lin, and 24 others. 2025. Qwen2.5 technical report. Preprint, arXiv:2412.15… find
  40. Rafael Rafailov, Archit Sharma, Eric Mitchell, Christopher D Manning, Stefano Ermon, and Chelsea Finn. 2023. Direct preference optimization: Your language model is secretly a reward model. Advances in neural information processing systems, 36:53728–53741.
  41. Rulin Shao, Shuyue Stella Li, Rui Xin, Scott Geng, Yiping Wang, Sewoong Oh, Simon Shaolei Du, Nathan Lambert, Sewon Min, Ranjay Krishna, Yulia Tsvetkov, Hannaneh Hajishirzi, Pang Wei Koh, and Luke Zettlemoyer. 2025. Spurious rewards: Rethinking training signals in rlvr. Preprint, arXiv:2506.10947. find
  42. Zhihong Shao, Peiyi Wang, Qihao Zhu, Runxin Xu, Junxiao Song, Xiao Bi, Haowei Zhang, Mingchuan Zhang, Y. K. Li, Y. Wu, and Daya Guo. 2024. Deepseekmath: Pushing the limits of mathematical reasoning in open language models. Preprint, arXiv:2402.03300.
  43. Guangming Sheng, Chi Zhang, Zilingfeng Ye, Xibin Wu, Wang Zhang, Ru Zhang, Yanghua Peng, Haibin Lin, and Chuan Wu. 2024. Hybridflow: A flexible and efficient rlhf framework. arXiv preprint arXiv: 2409.19256.
  44. Freda Shi, Mirac Suzgun, Markus Freitag, Xuezhi Wang, Suraj Srivats, Soroush Vosoughi, Hyung Won Chung, Yi Tay, Sebastian Ruder, Denny Zhou, Dipanjan Das, and Jason Wei. 2022. Language models are multilingual chain-of-thought reasoners. Preprint, arXiv:2210.03057.
  45. Shivalika Singh, Angelika Romanou, Clémentine Fourrier, David I. Adelani, Jian Gang Ngui, Daniel Vila-Suero, Peerat Limkonchotiwat, Kelly Marchisio, Wei Qi Leong, Yosephine Susanto, Raymond Ng, Shayne Longpre, Wei-Yin Ko, Madeline Smith, Antoine Bosselut, Alice Oh, Andre F. T. Martins, Leshem Choshe… — Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluation
  46. Guijin Son, Jiwoo Hong, Hyunwoo Ko, and James Thorne. 2025. Linguistic generalizability of test-time scaling in mathematical reasoning. arXiv preprint arXiv:2502.17407.
  47. Yuda Song, Dhruv Rohatgi, Aarti Singh, and J. Andrew Bagnell. 2025. To distill or decide? understanding the algorithmic trade-off in partially observable reinforcement learning. Preprint, arXiv:2510.03207. find
  48. Yuda Song, Gokul Swamy, Aarti Singh, J. Andrew Bagnell, and Wen Sun. 2024. The importance of online data: Understanding preference fine-tuning via coverage. In Advances in Neural Information Processing Systems, volume 37, pages 12243–12270. find
  49. Curran Associates, Inc. Gokul Swamy, Sanjiban Choudhury, J Bagnell, and Steven Z Wu. 2022. Sequence model imitation learning with unobserved contexts. Advances in Neural Information Processing Systems, 35:17665–17676. find
  50. Gokul Swamy, Sanjiban Choudhury, Wen Sun, Zhiwei Steven Wu, and J. Andrew Bagnell. 2025. All roads lead to likelihood: The value of reinforcement learning in fine-tuning. Preprint, arXiv:2503.01067. — [arXiv:2503.01067](https://arxiv.org/abs/2503.01067) find
  51. Gokul Swamy, Christoph Dann, Rahul Kidambi, Zhiwei Steven Wu, and Alekh Agarwal. 2024. A minimaximalist approach to reinforcement learning from human feedback. Preprint, arXiv:2401.04056. Zhi Rui Tam, Cheng-Kuang Wu, Yu Ying Chiu, Chieh-Yen Lin, Yun-Nung Chen, and Hung yi Lee. 2025. Language matters…
  52. Vladimir Vapnik and Akshay Vashist. 2009. A new learning paradigm: Learning using privileged information. Neural networks, 22(5-6):544–557. find
  53. Yi Xu, Laura Ruis, Tim Rocktäschel, and Robert Kirk. 2025. Investigating non-transitivity in llm-as-a-judge. Preprint, arXiv:2502.14074. — [arXiv:2502.14074](https://arxiv.org/abs/2502.14074) find
  54. 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…
  55. Seonghyeon Ye, Doyoung Kim, Sungdong Kim, Hyeonbin Hwang, Seungone Kim, Yongrae Jo, James Thorne, Juho Kim, and Minjoon Seo. 2024. Flask: Fine-grained language model evaluation based on alignment skill sets. Preprint, arXiv:2307.10928.
  56. Zheng-Xin Yong, M. Farid Adilazuarda, Jonibek Mansurov, Ruochen Zhang, Niklas Muennighoff, Carsten Eickhoff, Genta Indra Winata, Julia Kreutzer, Stephen H. Bach, and Alham Fikri Aji. 2025. Crosslingual reasoning through test-time scaling. Preprint, arXiv:2505.05408. — [arXiv:2505.05408](https://arxiv.org/abs/2505.05408) find
  57. A Yang Yue, Zhiqi Chen, Rui Lu, Andrew Zhao, Zhaokai Wang, Yang Yue, Shiji Song, and Gao Huang. 2025. Does reinforcement learning really incentivize reasoning capacity in llms beyond the base model? Preprint, arXiv:2504.13837.
  58. Weixiang Zhao, Jiahe Guo, Yang Deng, Tongtong Wu, Wenxuan Zhang, Yulin Hu, Xingyu Sui, Yanyan Zhao, Wanxiang Che, Bing Qin, Tat-Seng Chua, and Ting Liu. 2025. When less language is more: Language-reasoning disentanglement makes llms better multilingual reasoners. Preprint, arXiv:2505.15257. — [arXiv:2505.15257](https://arxiv.org/abs/2505.15257) find
  59. Lianmin Zheng, Wei-Lin Chiang, Ying Sheng, Siyuan Zhuang, Zhanghao Wu, Yonghao Zhuang, Zi Lin, Zhuohan Li, Dacheng Li, Eric P. Xing, Hao Zhang, Joseph E. Gonzalez, and Ion Stoica. 2023. Judging llm-as-a-judge with mt-bench and chatbot arena. Preprint, arXiv:2306.05685.
  60. Weihua Zheng, Xin Huang, Zhengyuan Liu, Tarun Kumar Vangani, Bowei Zou, Xiyan Tao, Yuhao Wu, Ai Ti Aw, Nancy F. Chen, and Roy Ka-Wei Lee. 2025. Adamcot: Rethinking cross-lingual factual reasoning through adaptive multilingual chain-of-thought. Preprint, arXiv:2501.16154. — [arXiv:2501.16154](https://arxiv.org/abs/2501.16154) find
  61. Jin Peng Zhou, Séb Arnold, Nan Ding, Kilian Q Weinberger, Nan Hua, and Fei Sha. 2025. Graders should cheat: privileged information enables expert-level automated evaluations. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 16583–16601.
  62. Wenhao Zhu, Shujian Huang, Fei Yuan, Shuaijie She, Jiajun Chen, and Alexandra Birch. 2024. Question translation training for better multilingual reasoning. In Findings of the Association for Computational Linguistics: ACL 2024, pages 8411–8423, Bangkok, Thailand. Association for Computational Lingui… find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sutawika-2026-gained-translation-privileged-pairwise-judges/