#agentic reinforcement learning

'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.'