跨语言的长链式推理
Long Chain-of-Thought Reasoning Across Languages
AI总结：
本研究探讨了跨语言长链式推理能力的扩展机制，发现模型规模扩展提升En-CoT性能，但Target-CoT表现滞后，尤其在数学推理中更为明显。通过预训练和后训练优化，多语言模型在推理效率和稳定性上取得进展，同时揭示了语言特定的失败模式。
摘要：
尽管大型推理模型在英语中展示了生成长链式推理（CoTs）的显著能力，但我们仍然缺乏理解这些长形式推理能力如何转移到大多数世界语言。在本工作中，我们系统地调查了模型开发的四个关键阶段--扩展、预训练、后训练和推理--以理解长CoT能力如何扩展到英语以外的语言。我们比较了两种推理设置，跨越九种非英语目标语言：En-CoT，其中模型处理目标语言输入，但用英语推理；以及Target-CoT，其中模型既处理输入又生成长CoTs。我们发现扩展推理模型的大小提升了En-CoT的多语言任务性能，但Target-CoT性能却滞后。这种差距在需要长、多步骤CoTs的任务如数学推理中更加明显。转向预训练，我们发现添加专门的推理阶段可以增强En-CoT性能但损害Target-CoT，而广谱多语言预训练同时提升两者。鉴于除英语外其他语言的高质量推理轨迹稀缺，我们探索了合成数据校准方法用于后训练。我们证明了在翻译自黄金英文轨迹的推理轨迹上微调优于在从大型推理模型中提取的目标语言轨迹上微调。最后，我们报告了不同语言之间的推理效率差异，并揭示了CoTs中的语言特定失败模式。我们发布模型、数据集和代码以促进进一步研究。
英文摘要：
While large reasoning models have shown remarkable ability to generate long chains-of-thought (CoTs) in English, we still lack understanding of how these long-form reasoning abilities transfer to the vast majority of the world's languages. In this work, we systematically investigate four key stages of model development--scaling, pretraining, post-training, and inference--to understand how long CoT capabilities extend beyond English. We compare two reasoning settings across nine non-English target languages: En-CoT, where models process target-language inputs, but reason in English; and Target-CoT, where models both process inputs and generate long CoTs in the target language. We find that scaling reasoning model size improves multilingual task performance in En-CoT, but Target-CoT performance lags behind. This gap widens for tasks requiring long, multi-step CoTs such as mathematical reasoning. Shifting to pretraining, we find that adding a specialized reasoning stage enhances En-CoT performance but degrades Target-CoT, whereas broad multilingual pretraining improves both modes simultaneously. Given the scarcity of high-quality reasoning traces in languages other than English, we explore synthetic data curation approaches for post-training. We demonstrate that fine-tuning on reasoning traces automatically translated from gold English traces outperforms fine-tuning on target-language traces distilled from large reasoning models. Finally, we report disparities in inference efficiency between languages and uncover language-specific failure modes in CoTs. We release models, datasets, and code to foster further research.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
