arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.20168cs.CL
收藏

DeepWideSearch:智能体信息检索中深度与广度的基准评测

DeepWideSearch: Benchmarking Depth and Width in Agentic Information Seeking

Alibaba International Digital Commerce(阿里巴巴国际数字商务)

Tian Lan, Bin Zhu, Qianghuai Jia, Junyang Ren, Haijun Li, Longyue Wang, Zhao Xu, Weihua Luo, Kaifu Zhang

AI总结:

针对现有搜索智能体无法兼顾多跳检索深度与信息收集广度的问题,提出首个基准DeepWideSearch,包含220个跨15领域的问题,实验显示最优智能体平均成功率仅2.39%,并识别出四种失败模式,以推动更强大的信息检索智能体研究。

摘要:

当前搜索智能体从根本上缺乏同时执行多跳检索的深度推理和大规模信息收集的广度的能力——这对于市场综合分析、商业拓展等现实应用而言是一个关键缺陷。为弥合这一差距,我们提出了DeepWideSearch,这是首个专门用于评估智能体在信息检索中整合深度与广度的基准。在DeepWideSearch中,智能体必须处理大量数据,每条数据都需要对多跳检索路径进行深度推理。具体而言,我们提出了两种方法将已有数据集进行转换,最终构建了一个涵盖15个不同领域、包含220个问题的精选集合。大量实验表明,即便最先进的智能体在DeepWideSearch上的平均成功率也仅为2.39%,这凸显了在信息检索任务中整合深度与广度搜索所面临的巨大挑战。此外,我们的错误分析揭示了四种失败模式:缺乏反思、过度依赖内部知识、检索不足以及上下文溢出——这些暴露了当前智能体架构的关键局限。我们公开发布DeepWideSearch,以促进未来对更强大、更稳健的信息检索智能体的研究。

英文摘要:

Current search agents fundamentally lack the ability to simultaneously perform \textit{deep} reasoning over multi-hop retrieval and \textit{wide}-scale information collection-a critical deficiency for real-world applications like comprehensive market analysis and business development. To bridge this gap, we introduce DeepWideSearch, the first benchmark explicitly designed to evaluate agents to integrate depth and width in information seeking. In DeepWideSearch, agents must process a large volume of data, each requiring deep reasoning over multi-hop retrieval paths. Specifically, we propose two methods to converse established datasets, resulting in a curated collection of 220 questions spanning 15 diverse domains. Extensive experiments demonstrate that even state-of-the-art agents achieve only 2.39% average success rate on DeepWideSearch, highlighting the substantial challenge of integrating depth and width search in information-seeking tasks. Furthermore, our error analysis reveals four failure modes: lack of reflection, overreliance on internal knowledge, insufficient retrieval, and context overflow-exposing key limitations in current agent architectures. We publicly release DeepWideSearch to catalyze future research on more capable and robust information-seeking agents.

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。

↑