@inproceedings{xu-etal-2026-fusesearch,
    title = "{F}use{S}earch: Learning Adaptive Parallel Execution for Efficient Code Localization",
    author = "Xu, Ke  and
      Xiao, Siyang  and
      Liang, Ming  and
      Yu, Yichen  and
      Wang, Zhixiang  and
      Xu, Jingxuan  and
      Chen, Dajun  and
      Jiang, Wei  and
      Li, Yong",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.findings-acl.143/",
    pages = "2954--2967",
    ISBN = "979-8-89176-395-1",
    abstract = "Code localization is a primary bottleneck in automated software development. While parallel tool execution can accelerate discovery, existing agents suffer from a 34.9{\%} redundant tool invocation rate, negating the benefits of parallelism. We introduce FuseSearch, which reframes parallel code localization as a quality{--}efficiency co-optimization problem. By defining tool efficiency{---}the ratio of novel information gain to total invocations{---}we employ a two-stage SFT and RL pipeline to train models in adaptive parallel strategies. Unlike fixed-breadth methods, FuseSearch dynamically adjusts search breadth based on task context, transitioning from exploration to refinement. On SWE-bench Verified, FuseSearch-4B matches SOTA performance (84.7{\%} file-level and 56.4{\%} function-level $F_1$ scores) while being 93.6{\%} faster, using 67.7{\%} fewer turns and 68.9{\%} fewer tokens. Our findings demonstrate that efficiency-aware training inherently boosts quality by eliminating noisy, redundant signals, enabling high-performance, low-cost localization agents. Code: \url{https://github.com/sxthunder/FuseSearch}"
}