The September 26 2025 research paper introduces FastGRPO, a high-efficiency framework designed to accelerate the training of large language models using Group Relative Policy Optimization. The authors identify that the generation phase is the primary bottleneck in reinforcement learning, accounting for over 90% of total training time. To solve this, they implement concurrency-aware speculative decoding, which dynamically adjusts drafting and verification strategies based on real-time batch sizes. Additionally, an online draft learning mechanism is introduced to keep the smaller assistant model aligned with the evolving target model. Experimental results show that this approach achieves end-to-end speedups of up to 2.72x without compromising reasoning performance. Ultimately, the framework optimizes hardware utilization by balancing memory bandwidth and computational overhead during high-concurrency training. September 26, 2026 Yizhou Zhang Ning Lv, Teng Wang, Jisheng Dang Lanzhou University, The University of Hong Kong, National University of Singapore https://arxiv.org/pdf/2509.21792