作者位次 #5/9OSDI 2026Oral· Operating Systems & Distributed Systems
RobustRL: Role-Based Fault Tolerance System for RL Post-Training
Zhenqian Chen · Baoquan Zhong · Xiang Li · Qing Dai · Xinkui Zhao · Miao Ye · Ren Cheng · Lufei Zhang · Jianwei Yin
摘要
RL post-training for LLMs has been widely scaled to enhance reasoning and tool-using capabilities. However, RL post-training interleaves training and inference workloads, exposing the system to faults from both sides. Existing fault tolerance frameworks for LLMs target either training or inferenc...