翻訳待ち：GRPO Beyond English: A Large-Scale Study of GRPO in Non-English and Multilingual Settings

AI サービスが一時的に利用できないため、復旧後に翻訳を補完します。

content type paperpublished August 2026 GRPO Beyond English: A Large-Scale Study of GRPO in Non-English and Multilingual Settings AuthorsKonstantin Dobler†, Federico Scozzafava, Jonathan Janke, Mohamed Ali, Simon Lehnerer View publication Reinforcement Learning with Verifiable Rewards (RLVR), often optimized with Group Relative Policy Optimization (GRPO), has become a central recipe for improving the reasoning capabilities of pretrained language models but current studies remain heavily English-centric. We conduct a large-scale empirical study of multilingual and non-English GRPO across a wide range of base models, training languages, and different reasoning language rewards. We find that training to reason in the native language often leaves only a small gap to training for English reasoning. We further observe strong crosslingual transfer: training in one language often improves performance in many others. However, specific trends are highly model- and language-dependent. In some cases, training in a particular language induces severe regressions on out-of-domain capabilities in other languages. Our analysis shows that RLVR beyond English can provide broad crosslingual gains, but also requires broad evaluation to detect language-specific regressions. † Hasso Plattner Institute & ELLIS Unit Potsdam Work done while at Apple Personalized Group Relative Policy Optimization for Heterogenous Preference Alignment April 2, 2026research area Methods and Algorithms, research area Speech and Natural Language Processing Despite their sophisticated general-purpose capabilities, Large Language Models (LLMs) often fail to align with diverse individual preferences because standard post-training methods, like Reinforcement Learning with Human Feedback (RLHF), optimize for a single, global objective. While Group Relative Policy Optimization (GRPO) is a widely adopted on-policy reinforcement learning framework, its group-based normalization implicitly assumes that all… Read more Do Large Language Models Have an English Accent? Evaluating and Improving the Naturalness of Multilingual LLMs May 16, 2025research area Speech and Natural Language Processingconference ACL Current Large Language Models (LLMs) are predominantly designed with English as the primary language, and even the few that are multilingual tend to exhibit strong English-centric biases. Much like speakers who might produce awkward expressions when learning a second language, LLMs often generate unnatural outputs in non-English languages, reflecting English-centric patterns in both vocabulary and grammar. Despite the importance of this issue,… Read more