The Reasoning Lingua Franca: A Double-Edged Sword for Multilingual AI
On the map: LLM reasoning → Reasoning benchmarks and evaluation
Abstract (arXiv)
Large Reasoning Models (LRMs) achieve strong performance on mathematical, scientific, and other question-answering tasks, but their multilingual reasoning abilities remain underexplored. When presented with non-English questions, LRMs often default to reasoning in English, raising concerns about interpretability and the handling of linguistic and cultural nuances. We systematically compare an LRM's reasoning in English versus the language of the question. Our evaluation spans two tasks: MGSM and GPQA Diamond. Beyond measuring answer accuracy, we also analyze cognitive attributes in the reasoning traces. We find that English reasoning traces exhibit a substantially higher presence of these cognitive behaviors, and that reasoning in English generally yields higher final-answer accuracy, with the performance gap increasing as tasks become more complex. However, this English-centric strategy is susceptible to a key failure mode - getting "Lost in Translation," where translation steps lead to errors that would have been avoided by reasoning in the language of the question.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 13 of its 23 references that are in the map. First posted on arXiv 23 Oct 2025.
Builds on
13 of 23 references.
Closest ancestors · 5
Same direction, newest first
- 2025When Models Reason in Your Language: Controlling Thinking Language Comes at the Cost of Accuracy28 May
- 2025Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?23 May
- 2025Qwen3 Technical Report14 May
- 2025Crosslingual Reasoning through Test-Time Scaling8 May
- 2023LLaMA: Open and Efficient Foundation Language Models27 Feb
Other references · 8
- 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,146
- 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
- 2024Gemma 2: Improving Open Language Models at a Practical Sizecited by 1,928
- 2024OpenAI o1 System Cardcited by 1,220
- 2022Language Models are Multilingual Chain-of-Thought Reasonerscited by 518
- 2025Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRscited by 373
- 2025PolyMath: Evaluating Mathematical Reasoning in Multilingual Contextscited by 52
- 2025Language Models' Factuality Depends on the Language of Inquirycited by 14
Built on by
Citing papers per year: 2026: 2.
2 papers.
Similar, not cited
Close in wording, no citation either way
- 2026What Makes Good Multilingual Reasoning? Disentangling Traces with Measurable Features
- 2026Rethinking the Multilingual Reasoning Gap with Layer Swap
- 2025Could Thinking Multilingually Empower LLM Reasoning?
- 2026Large Reasoning Models Are (Not Yet) Multilingual Latent Reasoners
- 2025Learn Globally, Speak Locally: Bridging the Gaps in Multilingual Reasoning
- 2024The Power of Question Translation Training in Multilingual Reasoning: Broadened Scope and Deepened Insights
- 2025A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across Languages
- 2025MultiNRC: A Challenging and Native Multilingual Reasoning Evaluation Benchmark for LLMs