mlsanityData to 14 Oct 2026

When Models Reason in Your Language: Controlling Thinking Language Comes at the Cost of Accuracy

arXiv 2505.22888 ↗First posted 28 May 2025cs.CLcited by 14

On the map: LLM reasoning → Reasoning benchmarks and evaluation

Abstract (arXiv)

Recent Large Reasoning Models (LRMs) with thinking traces have shown strong performance on English reasoning tasks. However, their ability to think in other languages is less studied. This capability is as important as answer accuracy for real world applications because users may find the reasoning trace useful for oversight only when it is expressed in their own language. We comprehensively evaluate two leading families of LRMs on our XReasoning benchmark and find that even the most advanced models often revert to English or produce fragmented reasoning in other languages, revealing a substantial gap in multilingual reasoning. Prompt based interventions that force models to reason in the users language improve readability and oversight but reduce answer accuracy, exposing an important trade off. We further show that targeted post training on just 100 examples mitigates this mismatch, though some accuracy loss remains. Our results highlight the limited multilingual reasoning capabilities of current LRMs and outline directions for future work. Code and data are available at https://github.com/Betswish/mCoT-XReasoning.

Key facts. Cited by 14 papers in the map (data through 14 Oct 2026). Placed in Reasoning benchmarks and evaluation, one of 17 directions in LLM reasoning. Builds on 28 of its 46 references that are in the map. First posted on arXiv 28 May 2025.

Builds on

28 of 46 references.

Closest ancestors · 4

Same direction, newest first

  1. 2025Crosslingual Reasoning through Test-Time Scaling8 May
  2. 2025A Survey of Multilingual Reasoning in Language Models13 Feb
  3. 2025Are DeepSeek R1 And Other Reasoning Models More Faithful?14 Jan
  4. 2023Not All Languages Are Created Equal in LLMs: Improving Multilingual Capability by Cross-Lingual-Thought Prompting11 May

Other references · 24

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  3. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learningcited by 6,330
  4. 2023GPQA: A Graduate-Level Google-Proof Q&A Benchmarkcited by 2,879
  5. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameterscited by 1,986
  6. 2025s1: Simple test-time scalingcited by 1,341
  7. 2024OpenAI o1 System Cardcited by 1,244
  8. 2025LIMO: Less is More for Reasoningcited by 498
  9. 2025Reasoning Models Don't Always Say What They Thinkcited by 415
  10. 2017Deep Reinforcement Learning: An Overviewcited by 333
  11. 2024Global MMLU: Understanding and Addressing Cultural and Linguistic Biases in Multilingual Evaluationcited by 191
  12. 2025Chain-of-Thought Reasoning In The Wild Is Not Always Faithfulcited by 170
12 more
  1. 2024From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Modelscited by 137
  2. 2025Effectively Controlling Reasoning Models through Thinking Interventioncited by 65
  3. 2024IrokoBench: A New Benchmark for African Languages in the Age of Large Language Modelscited by 43
  4. 2025Beyond Semantics: The Unreasonable Effectiveness of Reasonless Intermediate Tokenscited by 41
  5. 2024mHumanEval -- A Multilingual Benchmark to Evaluate Large Language Models for Code Generationcited by 32
  6. 2024Revisiting Weight Averaging for Model Mergingcited by 25
  7. 2024ProverbEval: Exploring LLM Evaluation Challenges for Low-resource Language Understandingcited by 15
  8. 2024Systematically Analyzing Prompt Injection Vulnerabilities in Diverse LLM Architecturescited by 12
  9. 2024Exploring Vulnerabilities and Protections in Large Language Models: A Surveycited by 11
  10. 2025On the Consistency of Multilingual Context Utilization in Retrieval-Augmented Generationcited by 10
  11. 2024Rethinking Weight-Averaged Model-mergingcited by 8
  12. 2023Comparative Analysis of Multilingual Text Classification & Identification through Deep Learning and Embedding Visualizationcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 6; 2026: 8.

14 papers.

Sort:

By area:

  1. 2025Pushing on Multilingual Reasoning Models with Language-Mixed Chain-of-Thoughtcited by 10
  2. 2025Evaluating Robustness of Large Language Models Against Multilingual Typographical Errorscited by 9
  3. 2025Beyond English-Centric Training: How Reinforcement Learning Improves Cross-Lingual Reasoning in LLMscited by 8
  4. 2026R3S: Refining and Recovering Reinforcement Signals for Multilingual Understanding and Reasoningcited by 5
  5. 2025Parallel Scaling Law: Unveiling Reasoning Generalization through A Cross-Linguistic Perspectivecited by 4
  6. 2026Language of Thought Shapes Output Diversity in Large Language Modelscited by 4
  7. 2025The Reasoning Lingua Franca: A Double-Edged Sword for Multilingual AIcited by 3
  8. 2026Rethinking the Multilingual Reasoning Gap with Layer Swapcited by 2
  9. 2026The Fragility of Chain-of-Thought Monitoring Across Typologically Diverse Languagescited by 1
  10. 2025Reveal-Bangla: A Dataset for Cross-Lingual Multi-Step Reasoning Evaluationcited by 0
  11. 2026mmPISA-bench: Do LLMs Reason Equally Well Across 43 Languages?cited by 0
  12. 2026When Does Reasoning Help in Machine Translation? A Hierarchical Analysis of LRM Reasoning Tracescited by 0
2 more
  1. 2026What Does Post-Training Change in Multilingual Reasoning?cited by 0
  2. 2026Investigating the Role of Reasoning-Language Alignment in Monolingual Retrieval-Augmented Generationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Language Matters: How Do Multilingual Input and Reasoning Paths Affect Large Reasoning Models?
  2. 2025Beg to Differ: Understanding Reasoning-Answer Misalignment Across Languages
  3. 2026What Makes Good Multilingual Reasoning? Disentangling Traces with Measurable Features
  4. 2026Building Multilingual Bridges: Data Mixing as the Pillar of Generalization for In-Language Reasoning
  5. 2025A Comprehensive Evaluation of Multilingual Chain-of-Thought Reasoning: Performance, Consistency, and Faithfulness Across Languages
  6. 2025Why Do Multilingual Reasoning Gaps Emerge in Reasoning Language Models?
  7. 2024Eliciting Better Multilingual Structured Reasoning from LLMs through Code
  8. 2025Language Mixing in Reasoning Language Models: Patterns, Impact, and Internal Causes