Elamur Research⌕
☰
home / papers / 2024 / LangBridge: Multilingual Reasoning Without Multilingual Supervision

LangBridge: Multilingual Reasoning Without Multilingual Supervision

Paper · 2024 · Dongkeun Yoon, Joel Jang, Sungdong Kim et al.

Summary

Reasoning-focused language models such as MetaMath and Orca 2 come from Llama 2 and are tuned on English domain data. As a result, they perform poorly on math, coding and logic problems written in low-resource languages. Adapting them to each language separately would require training corpora that often do not exist. LangBridge avoids this by connecting a multilingual encoder (the mT5 encoder) to a frozen reasoning LM through one trainable linear layer. The layer converts the encoder's final hidden states into soft prompts for the LM. Alignment uses only English data, with the original or closest available training set (e.g., MetaMathQA, OpenOrca). The models are then tested zero-shot on non-English inputs.

The approach clearly improves results for underrepresented languages across mathematical, code, logical and commonsense reasoning. Analysis with PCA and qualitative inspection links this to the largely language-neutral nature of mT5 representations. The method matters because it extends specialized English models to many languages without collecting any multilingual reasoning data. Code and models are publicly released.

Key points

  • MetaMath-13B combined with the mT5-XL encoder (2.2B) raises average MGSM accuracy from 40.5% to 53.5%. This slightly exceeds PaLM-540B, which scores 51.3%.
  • The setup is held fixed across experiments:
    • training data is capped at 200,000 English instances;
    • encoder inputs go up to 1,024 tokens and targets up to 128;
    • the LM stays frozen;
    • the encoder is frozen apart from its embeddings when adapting finetuned LMs, and trainable when adapting pretrained LMs.
  • A language counts as underrepresented if it makes up under 0.1% of Llama 2 pretraining data. Benchmarks include MGSM, translated HumanEval, Big-Bench Hard and XCOPA.
  • Chain-of-thought output stays mostly in English. In 178 Orca 2-LangBridge-15B rationales on the Bengali Snarks subtask, only eight showed accidental translation and seven contained Bengali. This suggests the representations are not fully language-agnostic.
  • There are two main limitations:
    • performance in high-resource languages sometimes drops;
    • a gap between high- and low-resource languages remains, depending on how well both the LM and the encoder handle each language.

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 States598 works · 191,225 citations🎓University of WashingtonSeattle · United States402 works · 212,762 citations🎓Korea Advanced Institute of Science and TechnologyDaejeon · South Korea56 works · 3,269 citations🏢NaverSeongnam-si · South Korea20 works · 12,146 citations🏢NAVER CloudSeongnam-si · South Korea16 works · 11,005 citations

 Fields, topics and concepts

📚 Shelves 5The 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 worksComputer Vision and Pattern Recognition · 3,486 works

🧭 Topics 2Research subjects and languages the work is about.Reasoning · 1,685 worksMultilingual & cross-lingual NLP · 882 works

Cite

Cite as Dongkeun Yoon, Joel Jang, Sungdong Kim, Seungone Kim, Sheikh Shafayat and Minjoon Seo. 2024. LangBridge: Multilingual Reasoning Without Multilingual Supervision. Proceedings of the 62nd 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/v1/2024.acl-long.405

@inproceedings{yoon2024langbridge,
  title = {{LangBridge: Multilingual Reasoning Without Multilingual Supervision}},
  author = {Dongkeun Yoon, Joel Jang, Sungdong Kim, Seungone Kim, Sheikh Shafayat and Minjoon Seo},
  booktitle = {Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  year = {2024},
  doi = {10.18653/v1/2024.acl-long.405},
  url = {https://doi.org/10.18653/v1/2024.acl-long.405}
}

References

67 references, 42 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Sweta Agrawal, Nikita Mehandru, Niloufar Salehi, and Marine Carpuat. 2022. Quality estimation via back-translation at the WMT 2022 quality estimation task. In Proceedings of the Seventh Conference on Machine Translation (WMT), pages 593–596, Abu Dhabi, United Arab Emirates (Hybrid). Association for… find
  2. Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Mensch, Katherine Millican, Malcolm Reynolds, et al. 2022. Flamingo: a visual language model for few-shot learning. Advances in Neural Information Processing Systems, 35:23716–23736.
  3. Zhangir Azerbayev, Hailey Schoelkopf, Keiran Paster, Marco Dos Santos, Stephen McAleer, Albert Q. Jiang, Jia Deng, Stella Biderman, and Sean Welleck. 2023. Llemma: An open language model for mathematics. arXiv preprint arXiv:2310.06786.
  4. Rachit Bansal, Bidisha Samanta, Siddharth Dalmia, Nitish Gupta, Sriram Ganapathy, Abhishek Bapna, Prateek Jain, and Partha Talukdar. 2024. LLM augmented LLMs: Expanding capabilities through composition. In The Twelfth International Conference on Learning Representations. find
  5. Rohan Bavishi, Erich Elsen, Curtis Hawthorne, Maxwell Nye, Augustus Odena, Arushi Somani, and Sağnak Taşırlar. 2023. Introducing our multimodal models. find
  6. Loubna Ben Allal, Niklas Muennighoff, Logesh Kumar Umapathi, Ben Lipkin, and Leandro von Werra. 2022. A framework for the evaluation of code generation models. https://github.com/bigcode-project/ bigcode-evaluation-harness. ↗
  7. Mark Chen, Jerry Tworek, Heewoo Jun, Qiming Yuan, Henrique Ponde de Oliveira Pinto, Jared Kaplan, Harri Edwards, Yuri Burda, Nicholas Joseph, Greg Brockman, et al. 2021. Evaluating large language models trained on code. arXiv preprint arXiv:2107.03374.
  8. Nuo Chen, Zinan Zheng, Ning Wu, Ming Gong, Yangqiu Song, Dongmei Zhang, and Jia Li. 2023. Breaking language barriers in multilingual mathematical reasoning: Insights and observations.
  9. Aakanksha Chowdhery, Sharan Narang, Jacob Devlin, Maarten Bosma, Gaurav Mishra, Adam Roberts, Paul Barham, Hyung Won Chung, Charles Sutton, Sebastian Gehrmann, et al. 2023. Palm: Scaling language modeling with pathways. Journal of Machine Learning Research, 24(240):1–113.
  10. Hyung Won Chung, Xavier Garcia, Adam Roberts, Yi Tay, Orhan Firat, Sharan Narang, and Noah Constant. 2023. Unimax: Fairer and more effective language sampling for large-scale multilingual pretraining. In The Eleventh International Conference on Learning Representations. find
  11. Karl Cobbe, Vineet Kosaraju, Mohammad Bavarian, Mark Chen, Heewoo Jun, Lukasz Kaiser, Matthias Plappert, Jerry Tworek, Jacob Hilton, Reiichiro Nakano, Christopher Hesse, and John Schulman. 2021. Training verifiers to solve math word problems. arXiv preprint arXiv:2110.14168.
  12. 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 fo…
  13. Marta R Costa-jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, et al. 2022. No language left behind: Scaling human-centered machine translation. arXiv preprint arXiv:2207.04672. find
  14. Fangxiaoyu Feng, Yinfei Yang, Daniel Cer, Naveen Arivazhagan, and Wei Wang. 2022. Language-agnostic BERT sentence embedding. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 878–891, Dublin, Ireland. Association for Computation…
  15. Jack FitzGerald, Christopher Hench, Charith Peris, Scott Mackie, Kay Rottmann, Ana Sanchez, Aaron Nash, Liam Urbach, Vishesh Kakarala, Richa Singh, Swetha Ranganath, Laurie Crist, Misha Britan, Wouter Leeuwis, Gokhan Tur, and Prem Natarajan. 2023. MASSIVE: A 1M-example multilingual natural language… find
  16. Leo Gao, Jonathan Tow, Baber Abbasi, Stella Biderman, Sid Black, Anthony DiPofi, Charles Foster, Laurence Golding, Jeffrey Hsu, Alain Le Noac’h, Haonan Li, Kyle McDonell, Niklas Muennighoff, Chris Ociepa, Jason Phang, Laria Reynolds, Hailey Schoelkopf, Aviya Skowron, Lintang Sutawika, Eric Tang, Ani… find
  17. Naman Goyal, Cynthia Gao, Vishrav Chaudhary, Peng-Jen Chen, Guillaume Wenzek, Da Ju, Sanjana Krishnan, Marc’Aurelio Ranzato, Francisco Guzmán, and Angela Fan. 2022. The Flores-101 evaluation benchmark for low-resource and multilingual machine translation. Transactions of the Association for Computat… find
  18. Dan Hendrycks, Collin Burns, Steven Basart, Andy Zou, Mantas Mazeika, Dawn Song, and Jacob Steinhardt. 2021a. Measuring massive multitask language understanding. Proceedings of the International Conference on Learning Representations (ICLR).
  19. Dan Hendrycks, Collin Burns, Saurav Kadavath, Akul Arora, Steven Basart, Eric Tang, Dawn Song, and Jacob Steinhardt. 2021b. Measuring mathematical problem solving with the math dataset. NeurIPS.
  20. Albert Q. Jiang, Alexandre Sablayrolles, Arthur Mensch, Chris Bamford, Devendra Singh Chaplot, Diego de las Casas, Florian Bressand, Gianna Lengyel, Guillaume Lample, Lucile Saulnier, Lélio Renard Lavaud, Marie-Anne Lachaux, Pierre Stock, Teven Le Scao, Thibaut Lavril, Thomas Wang, Timothée Lacroix,… find
  21. Wenxiang Jiao, Wenxuan Wang, Jen tse Huang, Xing Wang, Shuming Shi, and Zhaopeng Tu. 2023. Is chatgpt a good translator? yes with gpt-4 as the engine.
  22. Nikhil Kandpal, Haikang Deng, Adam Roberts, Eric Wallace, and Colin Raffel. 2023. Large language models struggle to learn long-tail knowledge.
  23. Tannon Kew, Florian Schottmann, and Rico Sennrich. 2023. Turning english-centric llms into polyglots: How much multilinguality is needed? find
  24. Viet Dac Lai, Chien Van Nguyen, Nghia Trung Ngo, Thuat Nguyen, Franck Dernoncourt, Ryan A. Rossi, and Thien Huu Nguyen. 2023. Okapi: Instruction-tuned large language models in multiple languages with reinforcement learning from human feedback.
  25. Angeliki Lazaridou, Adhiguna Kuncoro, Elena Gribovskaya, Devang Agrawal, Adam Liska, Tayfun Terzi, Mai Gimenez, Cyprien de Masson d’Autume, Tomas Kocisky, Sebastian Ruder, Dani Yogatama, Kris Cao, Susannah Young, and Phil Blunsom. 2021. Mind the gap: Assessing temporal generalization in neural langu… find
  26. Brian Lester, Rami Al-Rfou, and Noah Constant. 2021. The power of scale for parameter-efficient prompt tuning. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing, pages 3045–3059, Online and Punta Cana, Dominican Republic. Association for Computational Linguist…
  27. Junnan Li, Dongxu Li, Silvio Savarese, and Steven Hoi. 2023a. Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models.
  28. Raymond Li, Loubna Ben Allal, Yangtian Zi, Niklas Muennighoff, Denis Kocetkov, Chenghao Mou, Marc Marone, Christopher Akiki, Jia Li, Jenny Chim, et al. 2023b. Starcoder: may the source be with you! arXiv preprint arXiv:2305.06161. find
  29. Tianjian Li and Kenton Murray. 2023. Why does zero-shot cross-lingual generation fail? an explanation and a solution. In Findings of the Association for Computational Linguistics: ACL 2023, pages 12461–12476, Toronto, Canada. Association for Computational Linguistics. find
  30. Wing Lian, Bleys Goodson, Eugene Pentland, Austin Cook, Chanvichet Vong, and "Teknium". 2023. Openorca: An open dataset of gpt augmented flan reasoning traces. https://https://huggingface.co/Open-Orca/OpenOrca. find
  31. Jindřich Libovický, Rudolf Rosa, and Alexander Fraser. 2020. On the language neutrality of pre-trained multilingual representations. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 1663–1674, Online. Association for Computational Linguistics.
  32. Xi Victoria Lin, Todor Mihaylov, Mikel Artetxe, Tianlu Wang, Shuohui Chen, Daniel Simig, Myle Ott, Naman Goyal, Shruti Bhosale, Jingfei Du, Ramakanth Pasunuru, Sam Shleifer, Punit Singh Koura, Vishrav Chaudhary, Brian O’Horo, Jeff Wang, Luke Zettlemoyer, Zornitsa Kozareva, Mona Diab, Veselin Stoyano… — Few-shot Learning with Multilingual Language Models
  33. Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. 2023a. Improved baselines with visual instruction tuning.
  34. Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. 2023b. Visual instruction tuning.
  35. Ilya Loshchilov and Frank Hutter. 2019. Decoupled weight decay regularization. find
  36. Kelly Marchisio, Patrick Lewis, Yihong Chen, and Mikel Artetxe. 2023. Mini-model adaptation: Efficiently extending pretrained models to new languages via aligned shallow training. In Findings of the Association for Computational Linguistics: ACL 2023, pages 5474–5490, Toronto, Canada. Association fo… find
  37. Jack Merullo, Louis Castricato, Carsten Eickhoff, and Ellie Pavlick. 2023. Linearly mapping from image to text space. find
  38. Arindam Mitra, Luciano Del Corro, Shweti Mahajan, Andres Codas, Clarisse Simoes, Sahaj Agarwal, Xuxi Chen, Anastasia Razdaibiedina, Erik Jones, Kriti Aggarwal, Hamid Palangi, Guoqing Zheng, Corby Rosset, Hamed Khanpour, and Ahmed Awadallah. 2023. Orca 2: Teaching small language models how to reason.
  39. Subhabrata Mukherjee, Arindam Mitra, Ganesh Jawahar, Sahaj Agarwal, Hamid Palangi, and Ahmed Awadallah. 2023. Orca: Progressive learning from complex explanation traces of gpt-4.
  40. Miyu Oba, Tatsuki Kuribayashi, Hiroki Ouchi, and Taro Watanabe. 2023. Second language acquisition of neural language models. In Findings of the Association for Computational Linguistics: ACL 2023, pages 13557–13572, Toronto, Canada. Association for Computational Linguistics. find
  41. OpenAI. 2023. Gpt-4 technical report.
  42. 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…
  43. Arkil Patel, Satwik Bhattamishra, and Navin Goyal. 2021. Are NLP models really able to solve simple math word problems? In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2080–2094, Online. Associat…
  44. Telmo Pires, Eva Schlinger, and Dan Garrette. 2019. How multilingual is multilingual BERT? In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4996–5001, Florence, Italy. Association for Computational Linguistics.
  45. Edoardo Maria Ponti, Goran Glavaš, Olga Majewska, Qianchu Liu, Ivan Vulić, and Anna Korhonen. 2020. XCOPA: A multilingual dataset for causal commonsense reasoning. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 2362–2376, Online. Association…
  46. Maja Popović. 2015. chrF: character n-gram F-score for automatic MT evaluation. In Proceedings of the Tenth Workshop on Statistical Machine Translation, pages 392–395, Lisbon, Portugal. Association for Computational Linguistics.
  47. Libo Qin, Qiguang Chen, Fuxuan Wei, Shijue Huang, and Wanxiang Che. 2023. Cross-lingual prompting: Improving zero-shot chain-of-thought reasoning across languages. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 2695–2709.
  48. 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. Journal of Machine Learning Research, 21(140):1–67.
  49. Nils Reimers and Iryna Gurevych. 2020. Making monolingual sentence embeddings multilingual using knowledge distillation. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 4512–4525, Online. Association for Computational Linguistics.
  50. Melissa Roemmele, Cosmin Adrian Bejan, and Andrew S Gordon. 2011. Choice of plausible alternatives: An evaluation of commonsense causal reasoning. In 2011 AAAI Spring Symposium Series. find
  51. Baptiste Rozière, Jonas Gehring, Fabian Gloeckle, Sten Sootla, Itai Gat, Xiaoqing Ellen Tan, Yossi Adi, Jingyu Liu, Tal Remez, Jérémy Rapin, Artyom Kozhevnikov, Ivan Evtimov, Joanna Bitton, Manish Bhatt, Cristian Canton Ferrer, Aaron Grattafiori, Wenhan Xiong, Alexandre Défossez, Jade Copet, Faisal… — Code Llama: Open Foundation Models for Code
  52. Teven Le Scao, Angela Fan, Christopher Akiki, Ellie Pavlick, Suzana Ilić, Daniel Hesslow, Roman Castagné, Alexandra Sasha Luccioni, François Yvon, et al. 2022. Bloom: A 176b-parameter open-access multilingual language model. arXiv preprint arXiv:2211.05100. find
  53. Sheikh Shafayat, H M Quamran Hasan, Minhajur Rahman Chowdhury Mahim, Rifki Afina Putri, James Thorne, and Alice Oh. 2024. Benqa: A question answering and reasoning benchmark for bengali and english. find
  54. Peter Shaw, Jakob Uszkoreit, and Ashish Vaswani. 2018. Self-attention with relative position representations. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers), pages 464–468, New…
  55. 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. 2023. Language models are multilingual chain-of-thought reasoners. In The Eleventh International Conference on Learning Representa…
  56. Daria Soboleva, Faisal Al-Khateeb, Robert Myers, Jacob R Steeves, Joel Hestness, and Nolan Dey. 2023. SlimPajama: A 627B token cleaned and deduplicated version of RedPajama. find
  57. Mirac Suzgun, Nathan Scales, Nathanael Schärli, Sebastian Gehrmann, Yi Tay, Hyung Won Chung, Aakanksha Chowdhery, Quoc Le, Ed Chi, Denny Zhou, and Jason Wei. 2023. Challenging BIG-bench tasks and whether chain-of-thought can solve them. In Findings of the Association for Computational Linguistics: A…
  58. Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. 2022. Chain-of-thought prompting elicits reasoning in large language models. Advances in Neural Information Processing Systems, 35:24824–24837.
  59. 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…
  60. Longhui Yu, Weisen Jiang, Han Shi, Jincheng Yu, Zhengying Liu, Yu Zhang, James T. Kwok, Zhenguo Li, Adrian Weller, and Weiyang Liu. 2023. Metamath: Bootstrap your own mathematical questions for large language models.
  61. Zheng Yuan, Hongyi Yuan, Chengpeng Li, Guanting Dong, Keming Lu, Chuanqi Tan, Chang Zhou, and Jingren Zhou. 2023. Scaling relationship on learning mathematical reasoning with large language models.
  62. Wenhao Zhu, Yunzhe Lv, Qingxiu Dong, Fei Yuan, Jingjing Xu, Shujian Huang, Lingpeng Kong, Jiajun Chen, and Lei Li. 2023. Extrapolating large language models to non-english by aligning languages.
  63. Terry Yue Zhuo, Qiongkai Xu, Xuanli He, and Trevor Cohn. 2023. Rethinking round-trip translation for machine translation evaluation. In Findings of the Association for Computational Linguistics: ACL 2023, pages 319–337, Toronto, Canada. Association for Computational Linguistics. find
  64. Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, Aurelien Rodriguez, Armand Joulin, Edouard Grave, and Guillaume Lample. 2023a. Llama: Open and efficient foundation language models.
  65. Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yasmine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhosale, et al. 2023b. Llama 2: Open foundation and fine-tuned chat models. arXiv preprint arXiv:2307.09288.
  66. Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N Gomez, Łukasz Kaiser, and Illia Polosukhin. 2017. Attention is all you need. Advances in neural information processing systems, 30.
  67. Tu Vu, Aditya Barua, Brian Lester, Daniel Cer, Mohit Iyyer, and Noah Constant. 2022. Overcoming find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/yoon-2024-langbridge-multilingual-reasoning-without-multilingual/