arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2607.07548cs.CL
收藏

想得大,搜得小:分层搜索智能体中容量的关键所在?

Think Big, Search Small: Where Capacity Matters in Hierarchical Search Agents?

School of Data Science and Engineering(数据科学与工程学院)

Qinnan Cai, Yibo Zhao, Xiang Li

AI总结:

研究基于大语言模型的分层搜索智能体中模型容量分配问题,通过角色分解和受控容量扫描实验,发现容量敏感性不对称,分解是瓶颈,还给出了集中容量于委托、缩减执行容量且不牺牲精度的构建分层搜索智能体方法。

摘要:

基于大语言模型的搜索智能体越来越多地采用多智能体架构,主智能体将复杂问题分解为子查询并分发给并行的子智能体。然而,现有系统用相同规模的单一模型实例化所有角色,模型容量如何在各角色间分配仍未明确。我们将分层搜索分解为三个角色:负责任务分解的委托角色、负责检索和证据提取的执行角色、作为混淆控制保持固定的答案生成角色。然后在五个多跳问答基准上沿委托和执行轴进行受控容量扫描。实验有三个发现:一是角色分解始终优于单智能体基线;二是容量敏感性不对称,分解是能力瓶颈;三是通过质量过滤轨迹蒸馏训练的17亿参数执行器在精度上与前沿子智能体匹配且消耗的子智能体令牌少37%。这些结果为构建分层搜索智能体提供了具体方法。

英文摘要:

Large language model based search agents increasingly adopt multi-agent architectures in which a main agent decomposes a complex question into sub-queries and dispatches them to parallel sub-agents. However, existing systems instantiate all roles from a single model of identical scale, leaving open how model capacity should be distributed across roles. We factorize hierarchical search into three roles: a delegation role responsible for task decomposition, an execution role responsible for retrieval and evidence extraction, and an answer generation role held fixed as a confound control. We then conduct controlled capacity sweeps along the delegation and execution axes on five multi-hop QA benchmarks. The experiments yield three findings. First, role factorization consistently outperforms a single-agent baseline, improving exact match from 4.5 to 8.6 points across six model scales. Second, capacity sensitivity is asymmetric: scaling the delegation backbone improves EM by ~11 points, whereas scaling the execution sub-agent moves EM by only ~2.6 points, identifying decomposition as the capability bottleneck. Third, a 1.7B-parameter executor trained via quality-filtered trajectory distillation matches a frontier sub-agent in accuracy while consuming 37% fewer sub-agent tokens, advancing the Pareto frontier. These results suggest a concrete recipe for building hierarchical search agents: concentrate capacity at delegation and downsize execution without sacrificing accuracy. Our code is available at https://github.com/QinnanCai0115/role-factorized-search.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑