Apple's large-scale study: GRPO reasoning in native languages nearly matches English
Apple ML Research · rss · 2026-08-18
Apple ML Research publishes "GRPO Beyond English," a large-scale empirical study of multilingual and non-English GRPO, addressing how English-centric current RLVR research is. The study spans a wide range of base models, training languages, and different reasoning-language rewards.
Key finding: training models to reason in their native language often leaves only a small gap compared to training for English reasoning—suggesting reasoning RL can be done effectively in local languages.
More from Research
- Transformer trained on flying bat recordings yields hippocampus 'digital twin' revealing hidden structure — patrickmineault · 2026-10-16
- One year of Recursive Language Models: GPT-5-mini beats GPT-5 by 110% on 132k-token tasks, cheaper too — a1zhang · 2026-10-16
- MIT PhD Sharut Gupta joins DeepMind; Isola hails her representational alignment defense talk — phillip_isola · 2026-10-16
- 36 primitives, just 10 parameters each: a remarkably precise T-pose body fit — letandrewcook · 2026-10-16
- Interp researcher pushes back: if you can't simulate a neuron, you haven't explained it — nabla_theta · 2026-10-16
- MLP-Splatting (ECCV 2026) code released: MLPs as light-field primitives instead of Gaussians — AjdDavison · 2026-10-16