IsoCompute Playbook: Optimally Scaling Sampling Compute for LLM RL
UC San Diego · University of California, San Diego · Carnegie Mellon University · MBZUAI Institute of Foundational Models · Institute of Foundation Models, Mohamed bin Zayed University of Artificial Intelligence · Petuum Inc. and CMU · CMU · Lila Sciences · Indian Institute of Technology Bombay
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
While scaling laws guide compute allocation for LLM pre-training, analogous prescriptions for reinforcement learning (RL) post-training of LLMs remain poorly understood. We study the compute-optimal allocation of sampling compute for on-policy RL methods in LLMs, framing scaling as a compute-constrained optimization over three resources: parallel rollouts per problem, number of problems per batch, and number of update steps. We find that the compute-optimal number of parallel rollouts per problem increases predictably with compute budget and then saturates. This trend holds across both easy and hard problems, though driven by different mechanisms: solution sharpening on easy problems and coverage expansion on hard problems. We further show that increasing the number of parallel rollouts mitigates interference across problems, while the number of problems per batch primarily affects training stability and can be chosen within a broad range. Validated across base models and data distributions, our results recast RL scaling laws as prescriptive allocation rules and provide practical guidance for compute-efficient LLM RL post-training.