Scaling Test-time Compute for Low-resource Languages: Multilingual Reasoning in LLMs (2504.02890v1)

Abstract: Recent advances in test-time compute scaling have enabled LLMs to tackle deep reasoning tasks by generating a chain-of-thought (CoT) that includes trial and error, backtracking, and intermediate reasoning steps before producing the final answer. However, these techniques have been applied predominantly to popular languages, such as English, leaving reasoning in low-resource languages underexplored and misaligned. In this work, we investigate the multilingual mechanism by which LLMs internally operate in a latent space biased toward their inherently dominant language. To leverage this phenomenon for low-resource languages, we train models to generate the CoT in English while outputting the final response in the target language, given input in the low-resource language. Our experiments demonstrate that this approach, named English-Pivoted CoT Training, outperforms other baselines, including training to generate both the CoT and the final response solely in the target language, with up to 28.33% improvement. Further analysis provides novel insights into the relationships between reasoning and multilinguality of LLMs, prompting for better approaches in developing multilingual large reasoning models