@inproceedings{zhang-etal-2025-hierarchical,
    title = "Hierarchical Reward Modeling for Fault Localization in Large Code Repositories",
    author = "Zhang, Jiwei  and
      Lian, Jianxun  and
      Qin, Haiming  and
      Zhou, Mingyang  and
      Lu, KeZhong  and
      Mao, Rui  and
      Liao, Hao",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.findings-emnlp.966/",
    doi = "10.18653/v1/2025.findings-emnlp.966",
    pages = "17782--17796",
    ISBN = "979-8-89176-335-7",
    abstract = "Large Language Models (LLMs) exhibit significant potential in complex software engineering tasks, however, their fault localization capabilities within repository are constrained by inherent limitations in max context length. Although Test-Time Scaling (TTS) can generate multiple candidate solutions, traditional selection strategies often fail to identify the optimal one. To solve this problem, we introduces \textbf{Hi}erarchical \textbf{Lo}calization \textbf{R}eward \textbf{M}odel (HiLoRM), which specifically designed to evaluate and select the most accurate fault localization candidates (at file, function, and line levels) from the multiple sampled outputs of LLMs, thereby enhancing localization accuracy. Furthermore, we constructed the HiFL-44k dataset, comprising approximately 44,000 fault localization instances, to train HiLoRM. Experimental results demonstrate that on the SWE-Bench-Lite dataset, HiLoRM improves the final line-level localization recall by 12{\%} compared to a baseline model that does not use a reward model. Concurrently, HiLoRM exhibits a strong capability to evaluate predictions from larger LLMs (e.g., 32B parameters) and demonstrates transferability and generalization potential when applied to other fault localization methods. This work provides an effective methodology and an accessible model to significantly improve the accuracy and reliability of LLMs for repository-level fault localization. Our codes and datasets are available at https://github.com/SZU-ZJW/HiFL-Method."
}