#math reasoning

'MobileLLM-R1 is Meta's compact edge reasoning family (140M–950M parameters) that achieves significant math and coding gains over other open models while being far more data efficient.'

'Microsoft's rStar2-Agent integrates code execution into the reasoning loop, allowing a 14B model to outperform larger systems on math benchmarks with shorter reasoning traces.'

GURU introduces a multi-domain reinforcement learning dataset and models that significantly improve reasoning abilities of large language models across six diverse domains, outperforming previous open models.

AREAL is a new asynchronous reinforcement learning system that significantly speeds up training of large reasoning models by separating generation and training processes, achieving up to 2.77× faster training without loss of accuracy.

Qwen2.5-Math models improve math reasoning significantly even when trained with incorrect or random reward signals, highlighting unique reinforcement learning dynamics not seen in other models.