arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.08049cs.CLcs.AI
收藏

对过程奖励模型的调查:从结果信号到大语言模型的过程监督

A Survey of Process Reward Models: From Outcome Signals to Process Supervisions for Large Language Models

Shanghai Jiao Tong University(上海交通大学);University College London(伦敦大学学院);Carnegie Mellon University(卡内基梅隆大学);University of Bristol(布里斯托大学)

Congmin Zheng, Jiachen Zhu, Zhuoying Ou, Yuxiang Chen, Kangning Zhang, Rong Shan, Zeyu Zheng, Mengyue Yang, Jianghao Lin, Yong Yu, Weinan Zhang

AI总结:

本文综述了过程奖励模型,探讨了如何生成过程数据、构建PRMs及在测试时扩展和强化学习中的应用,涵盖数学、代码、文本、多模态推理、机器人和智能体等领域。

摘要:

尽管大语言模型(LLMs)展现出先进的推理能力,传统对齐仍主要依赖仅判断最终答案的结果奖励模型(ORMs)。过程奖励模型(PRMs)通过在步骤或轨迹层面评估和引导推理来弥补这一差距。本文通过完整循环系统地回顾了PRMs:如何生成过程数据、构建PRMs以及在测试时扩展和强化学习中使用PRMs。我们总结了数学、代码、文本、多模态推理、机器人和智能体等领域的应用,并回顾了新兴基准。我们的目标是澄清设计空间,揭示开放挑战,并引导未来研究朝着细粒度、稳健的推理对齐方向发展。

英文摘要:

Although Large Language Models (LLMs) exhibit advanced reasoning ability, conventional alignment remains largely dominated by outcome reward models (ORMs) that judge only final answers. Process Reward Models(PRMs) address this gap by evaluating and guiding reasoning at the step or trajectory level. This survey provides a systematic overview of PRMs through the full loop: how to generate process data, build PRMs, and use PRMs for test-time scaling and reinforcement learning. We summarize applications across math, code, text, multimodal reasoning, robotics, and agents, and review emerging benchmarks. Our goal is to clarify design spaces, reveal open challenges, and guide future research toward fine-grained, robust reasoning alignment.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑