表征对话式LLM智能体的网络搜索：从搜索决策与策略到结果与响应

中文摘要 AI 辅助

对话式LLM智能体日益依赖网络搜索，然而智能体搜索的端到端生命周期仍鲜为人知。我们首次对四个主要对话平台（ChatGPT、Claude、Grok和DeepSeek）上的网络搜索进行了研究，将真实用户交互（in-vivo）与使用同一平台模型通过其API进行的受控实验（in-vitro）相结合。我们考察了智能体调用网络搜索的决策质量、其制定查询的策略、其收到的搜索结果中潜在的领域偏好，以及其在将搜索结果转化为有依据的响应时所做出的选择。我们发现，网络搜索决策在不同平台和模型之间存在显著差异，而更频繁地调用网络搜索并不一定带来更好的响应质量。我们进一步表明，对话式智能体采用不同的复杂查询策略，且平台特定的搜索引擎会返回来自其偏好领域的搜索结果。最后，尽管响应在很大程度上基于搜索结果，但一些主张依赖于未引用的搜索结果，这引发了对归因和可靠性的担忧。我们的发现对未来AI智能体及为对话式检索优化的网络搜索工具的设计具有重要意义。

英文摘要

Conversational LLM agents increasingly rely on Web search, yet the end-to-end lifecycle of agentic search remains poorly understood. We present the first study of Web search across four major conversational platforms (ChatGPT, Claude, Grok, and DeepSeek), combining real-world user interactions (invivo) with controlled experiments using the same platform's models by their APIs (invitro). We investigate the quality of agentic decisions to invoke Web search, their strategies to formulate queries, the potential domain preferences in the search results they receive, and the choices they make when transforming search results into grounded responses. We find that Web-search decisions vary substantially across platforms and models, while more frequent Web-search invocation does not necessarily yield better response quality. We further show that conversational agents employ different complex querying strategies and that platform specific search engines return search results from their preferred domains. Finally, although responses are largely grounded in search results, some claims rely on uncited search results, raising concerns about attribution and reliability. Our findings have important implications for the design of future AI agents and Web search tools optimized for conversational retrieval.

发表机构

• Max Planck Institute for Software Systems（马克斯·普朗克软件系统研究所）
• Saarland University（萨尔大学）
• Laboratoire d’Informatique de l’École Polytechnique（巴黎综合理工学院计算机科学实验室）
• Seoul National University（首尔大学）
• Ruhr University Bochum（波鸿鲁尔大学）
• Microsoft（微软）

机构由 AI 辅助整理，请以论文原文为准。