@inproceedings{wang-etal-2026-prime,
    title = "{PRIME}: A Process-Outcome Alignment Benchmark for Verifiable Reasoning in Mathematics and Engineering",
    author = "Wang, Xiangfeng  and
      Guo, Hangyu  and
      Lai, Yanlin  and
      Huang, Mitt  and
      Zhao, Liang  and
      Yao, Chengyuan  and
      Zhang, Yinmin  and
      Han, Qi  and
      Xiaoxiaoren  and
      Yuan, Chun  and
      Xu, Tong  and
      Ge, Zheng  and
      Zhang, Xiangyu  and
      Jiang, Daxin",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.683/",
    pages = "14973--14985",
    ISBN = "979-8-89176-390-6",
    abstract = "While model-based verifiers are essential for scaling Reinforcement Learning with Verifiable Rewards (RLVR), current outcome-centric verification paradigms primarily focus on the consistency between the final result and the ground truth, often neglecting potential errors in the derivation process. This leads to assigning positive rewards to correct answers produced from incorrect derivations. To bridge this gap, we introduce **PRIME**, a benchmark for evaluating verifiers on **PR**ocess-outcome alignment verification **I**n **M**athematics and **E**ngineering. Curated from a comprehensive collection of college-level STEM problems, **PRIME** comprises 2,530 high-difficulty samples through a consistency-based filtering pipeline. Through extensive evaluation, we find that current verifiers frequently fail to detect derivation flaws. Furthermore, we propose a process-aware RLVR training paradigm utilizing verifiers selected via **PRIME**. This approach substantially outperforms the outcome-only verification baseline, achieving absolute performance gains of **8.29{\%}**, **9.12{\%}**, and **7.31{\%}** on AIME24, AIME25, and Beyond-AIME, respectively, for the Qwen3-14B-Base model. Finally, we demonstrate a strong linear correlation ($R^2 > 0.92$) between verifier accuracy on **PRIME** and RLVR training effectiveness, validating **PRIME** as a reliable predictor for verifier selection."
}