1
PlexRL: Cluster-Level Orchestration of Serviceized LLM Execution for RLVR
提出PlexRL框架，在集群层面编排服务化LLM执行以优化RLVR，显著提升推理效率与资源利用率
arXiv:2605.20863v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) has recently unlocked strong reasoning capabil…