Tournament-GRPO: Group-Wise Tournament Rewards for Reinforcement Learning in Open-Ended Long-Form Generation 文章

ArXiv CS.CL2026-05-27NEWSen作者: Zixuan Yang, Yiqun Chen, Wei Yang, Erhan Zhang, Zihan Shen, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Jiaxin Mao
查看原文 →
关系图谱

Tournament-GRPO: Group-Wise Tournament Rewards for Reinforcement Learning in Open-Ended Long-Form Generation · 相关技术

相关技术

ginORM远程代码执行（RCE）研究法辛烷值reinforcement learningdivide-and-conquer partitioningStraight-Through EstimatorQueryParts-of-Speech (POS) tagsNATMaliLMMLLM-as-a-JudgeGrouped Memorization EvaluationGroupGRPOForFFIANNdeep research