paper-with-me

Papers

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

2025-05-28 · Jirui Qi, Shan Chen, Zidi Xiong, Raquel Fernández, Danielle S. Bitterman, Arianna Bisazza

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.

📄 PDF Abstract BibTeX arXiv:2505.22888

Code (1)

betswish/mcot-xreasoning 공식 구현 pytorch

Similar Papers 제목 키워드 기반

Effectively Controlling Reasoning Models through Thinking Intervention

2025-03-31 · Tong Wu, Chong Xiang, Jiachen T. Wang, G. Edward Suh 외

Reasoning-enhanced large language models (LLMs) explicitly generate intermediate reasoning steps prior to generating final answers, helping the model excel in complex problem-solving. In this paper, we demonstrate that t…

Instruction FollowingSafety Alignment

Reasoning Models Can Be Effective Without Thinking

2025-04-14 · Wenjie Ma, Jingxuan He, Charlie Snell, Tyler Griggs 외

Recent LLMs have significantly improved reasoning capabilities, primarily by including an explicit, lengthy Thinking process as part of generation. In this paper, we question whether this explicit thinking is necessary. …

Automated Theorem ProvingMathematical Problem-Solving

Know When to Stop: Segment-Level Credit Assignment for Reducing Overthinking

2026-07-01 · Chia-Hsuan Lee, Sihui Dai, Mingyang Zhou, Isha Slavin 외 arxiv

Reasoning language models frequently overthink: generating extended chains of behaviors such as hedging, approach abandonment, and self contradiction that consume tokens without improving answers. We show that these beha…

Controlling Thinking Speed in Reasoning Models

2025-07-04 · Zhengkai Lin, Zhihang Fu, Ze Chen, Chao Chen 외 arxiv

Human cognition is theorized to operate in two modes: fast, intuitive System 1 thinking and slow, deliberate System 2 thinking. While current Large Reasoning Models (LRMs) excel at System 2 thinking, their inability to p…

Latent Thinking Optimization: Your Latent Reasoning Language Model Secretly Encodes Reward Signals in Its Latent Thoughts

2025-09-30 · Hanwen Du, Yuxin Dong, Xia Ning arxiv

Large Language Models (LLMs) excel at problem solving by generating chain of thoughts in natural language, but such verbal thinking is computationally costly and prone to overthinking. A recent work instead proposes a la…