大模型在多语言中存在隐式推理，但能力差异显著。
Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners
- 通过截断推理过程，发现模型在未完成文本推理时已能正确作答。
- 资源丰富语言的隐式推理更明显，困难任务上普遍较弱。
- 不同语言的内部推理路径高度一致，倾向英语主导模式。
大型推理模型（LRMs）在数学推理任务中表现优异，常归因于其生成显式思维链（CoT）的能力。然而近期研究发现，这些模型往往在完成文本推理前就已得出正确答案，表明存在隐式推理——即编码于隐藏状态中的非文本内部计算。尽管该现象已在英语中被探索，其多语言行为仍不明确。本文系统考察了11种语言中LRMs的多语言隐式推理。采用截断策略，分析模型仅获得部分推理痕迹时正确答案如何逐步形成，从而测量隐式预测的生成过程。结果揭示出明显的多语言隐式推理证据，但分布不均：资源丰富语言中较强，低资源语言中较弱，且在困难基准上普遍不明显。为进一步理解差异是否反映内部机制不同，我们进行表示分析。尽管表面差异明显，内部预测演化在各语言间高度一致，总体与英语模式趋同，暗示存在以英语为中心的隐式推理路径。
原文摘要 · Abstract (English)
Large reasoning models (LRMs) achieve strong performance on mathematical reasoning tasks, often attributed to their capability to generate explicit chain-of-thought (CoT) explanations. However, recent work shows that LRMs often arrive at the correct answer before completing these textual reasoning steps, indicating the presence of latent reasoning -- internal, non-verbal computation encoded in hidden states. While this phenomenon has been explored in English, its multilingual behavior remains largely unknown. In this paper, we conduct a systematic investigation of multilingual latent reasoning in LRMs across 11 languages. Using a truncation-based strategy, we examine how the correct answer emerges as the model is given only partial reasoning traces, allowing us to measure stepwise latent prediction formation. Our results reveal clear evidence of multilingual latent reasoning, though unevenly: strong in resource-rich languages, weaker in low-resource ones, and broadly less observable on harder benchmarks. To understand whether these differences reflect distinct internal mechanisms, we further perform representational analyses. Despite surface-level disparities, we find that the internal evolution of predictions is highly consistent across languages and broadly aligns with English -- a pattern suggesting an English-centered latent reasoning pathway.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。