作者位次 #3/14OSDI 2026Oral· Operating Systems & Distributed Systems
Weave: Efficient Co-Scheduling for Disaggregated RL Post-Training
Tianyuan Wu · Lunxi Cao · Yining Wei · Wei Gao · Yuheng Zhao · Dakai An · Shaopan Xiong · Zhiqiang Lv · Ju Huang · Siran Yang · Yinghao Yu · Jiamang Wang · Lin Qu · Wei Wang
摘要
Rollout–training disaggregation is emerging as the standard architecture for Reinforcement Learning (RL) post-training, where memory-bound rollout and compute-bound training are physically disaggregated onto purpose-built clusters to maximize hardware efficiency. However, the strict synchronizati...