#HARGO

1 stories · 0 in 7 days · 0 in 30 days

GRPO 1 HPC 1 RL后训练 1

Stories, products and papers tagged "HARGO", newest first.

Jul 31

Chinese contentHARGO：异构感知奖励引导优化用于LLM的HPC任务后训练

HARGO给HPC任务的RL后训练提了个新思路，不用标签就能按难度加权，四个任务上全面超过GRPO。

#HARGO #HPC #RL后训练 #GRPO

arXiv cs.LG@Tiangang Li, Xiangbo TianSource