cs.CLNov 16, 2026

GPRM: Global Perspective Process Reward Model

Authors: Zifan Zhang, Zhengbo Jiao, Shaobo Wang, Jing Zhou, Guoyi Li, Jiayi Chen, Wei Wang, Bing Zhao, 

Organizations: Wuhan University · Alibaba Group · Shanghai University of Finance and Economics · Shanghai Jiao Tong University · 

Abstract

Process Reward Models (PRMs) are widely used for step-level evaluation of multi-step reasoning in large language models, but existing methods are often limited by weak historical dependency modeling and insufficient global awareness. They typically fail to leverage prior step evaluations for coherent judgment and lack visibility into subsequent reasoning, leading to suboptimal assessment of intermediate steps. To address this issue, we propose the Global Perspective Process Reward Model (GPRM), which jointly models the problem, previous reasoning steps and their evaluations, the current step, and future steps to enable history-aware and globally informed process evaluation. We further introduce a structured evaluation framework along four dimensions---look back, look ahead, self-check, and goal alignment---to assess prior consistency, future plausibility, step validity, and task alignment. Based on this framework, we construct datasets for supervised fine-tuning (SFT) and reinforcement learning (GRPO). Experiments show that, despite using less training data and smaller model sizes, GPRM outperforms strong baseline models by 7.8 and 5.4 points on PRMBench and ProcessBench, respectively. GPRM also generalizes well across tasks, surpassing the GPT-4.1 baseline on Agent Error Bench, improving strong baselines by around 4 points under test-time scaling (TTS), and achieving competitive performance on PPE.

Figures & tables

Appendix figures & tables1 asset

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. The Bidirectional Process Reward Model

    Aug 3, 2025Lingyin Zhang, Jun Gao, Xiaoxue Ren +1Process Reward ModelsReasoning Verification

  2. Unsupervised Process Reward Models

    May 11, 2026Artyom Gadetsky, Maxim Kodryan, Siba Smarak Panigrahi +2Reward ModelingProcess Reward Models

  3. ScalePRM: Training Process Reward Models by Scaling Verification Compute Without Ground Truth

    Dec 2, 2025Salman Rahman, Sruthi Gorantla, Arpit Gupta +3Process Reward ModelsReinforcement Learning with Verifiable Rewards