作者位次 #24/29OSDI 2026Oral· Operating Systems & Distributed Systems
RLinf: Flexible and Efficient Large-Scale Reinforcement Learning via Macro-to-Micro Flow Transformation
Chao Yu · Yuanqing Wang · Zhen Guo · Hao Lin · Si Xu · Hongzhi Zang · Quanlu Zhang · Yongji Wu · Chunyang Zhu · Junhao Hu · Zixiao Huang · Mingjie Wei · Yuqing Xie · Ke Yang · Bo Dai · Zhexuan Xu · Jiakun Du · Xiangyuan Wang · Xu Fu · Letong Shi · Zhihao Liu · Kang Chen · Weilin Liu · Gang Liu · Boxun Li · Jianlei Yang · Tmall Group of Alibaba) · Guohao Dai · Yu Wang
摘要
Reinforcement learning (RL) has demonstrated immense potential in advancing artificial general intelligence, agentic intelligence, and embodied intelligence. However, the inherent heterogeneity and dynamicity of RL workflows often lead to low hardware utilization and slow training on existing sys...