SOURCE-LINKED INTELLIGENCE

When Do Larger Batches Help Scale LLM Reinforcement Learning?

arXiv · AI, language, vision and robotics · article · Aug 29, 2026 · UTC

Larger batches reduce the variance of stochastic gradients per update and are therefore often expected to accelerate training. Yet whether this statistical benefit translates into lower wall-clock time-to-target remains unclear, because each update consumes more samples and may take longer to execute. We study this tradeoff in reinforcement learning for large language models. We separate its algorithmic and systems effects by comparing learning and execution along their natural axes. At the algorithmic level, we compare configurations at equal cumulative sample counts while retuning batch-depe

Read original source ↗ Open in workspace

recordType

paper

region

Global