@inproceedings{chang-etal-2025-step,
    title = "Step-level Verifier-guided Hybrid Test-Time Scaling for Large Language Models",
    author = "Chang, Kaiyan  and
      Shi, Yonghao  and
      Wang, Chenglong  and
      Zhou, Hang  and
      Hu, Chi  and
      Liu, Xiaoqian  and
      Luo, Yingfeng  and
      Ge, Yuan  and
      Xiao, Tong  and
      Zhu, JingBo",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.931/",
    pages = "18473--18488",
    ISBN = "979-8-89176-332-6",
    abstract = "Test-Time Scaling (TTS) is a promising approach to progressively elicit the model{'}s intelligence during inference. Recently, training-based TTS methods, such as continued reinforcement learning (RL), have further surged in popularity, while training-free TTS methods are gradually fading from prominence. However, the additional computation overhead of training amplifies the burden on test-time scaling.In this paper, we focus on training-free TTS methods for reasoning. We first design Conditional Step-level Self-refinement, a fine-grained sequential scaling method guided by process verification. On top of its effectiveness, we further combine it with other classical parallel scaling methods at the step level, to introduce a novel inference paradigm called Hybrid Test-Time Scaling. Extensive experiments on five instruction-tuned LLMs across different scales (3B-14B) and families demonstrate that hybrid strategy incorporating various training-free TTS methods at a fine granularity has considerable potential for expanding the reasoning performance boundaries of LLMs."
}