@inproceedings{chen-etal-2025-search,
    title = "Search-in-Context: Efficient Multi-Hop {QA} over Long Contexts via {M}onte {C}arlo Tree Search with Dynamic {KV} Retrieval",
    author = "Chen, Jiabei  and
      Liu, Guang  and
      He, Shizhu  and
      Luo, Kun  and
      Xu, Yao  and
      Zhao, Jun  and
      Liu, Kang",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingestion-acl-25/2025.findings-acl.1356/",
    pages = "26443--26455",
    ISBN = "979-8-89176-256-5",
    abstract = "Recent advancements in large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks, such as math problem-solving and code generation. However, multi-hop question answering (MHQA) over long contexts, which demands both robust knowledge-intensive reasoning and efficient processing of lengthy documents, remains a significant challenge. Existing approaches often struggle to balance these requirements, either neglecting explicit reasoning or incurring expensive computational costs due to full-attention mechanisms over long contexts. To address this, we propose **Search-in-Context (SIC)**, a novel framework that integrates Monte Carlo Tree Search (MCTS) with dynamic key-value (KV) retrieval to enable iterative, context-aware reasoning. SIC dynamically retrieves critical KV pairs (e.g., 4K tokens) at each step, prioritizing relevant evidence while mitigating the ``lost in the middle'' problem. Furthermore, the paper introduces a Process-Reward Model (PRM) trained on auto-labeled data to guide the MCTS process with stepwise rewards, promoting high-quality reasoning trajectories without manual annotation. Experiments on three long-context MHQA benchmarks (HotpotQA, 2WikiMultihopQA, MuSiQue) and a counterfactual multi-hop dataset demonstrate SIC{'}s superiority, achieving state-of-the-art performance while significantly reducing computational overhead."
}