AI 中文总结
提出图推理智能体GRA，探索混合知识图谱，在工业基准UFK-M上准确率优于全上下文智能体，读取令牌更少，增益源于选择性访问。
AI 中文摘要
调用工具的大语言模型智能体通过少量通用原语（列出、读取和搜索文件：ls、cat、grep）来导航陌生代码库。知识图谱可提供相同的接口：列出邻居、读取节点内容和搜索描述是在不同载体上的相同操作。基于这种对应关系，我们提出GRA（Graph Reasoning Agent，图推理智能体），该智能体探索混合知识图谱（其节点为文本概念或关系表），使用7种通用工具，在运行时发现所有领域特定内容。在UFK-M（统一工厂知识模型，一个包含258个分析问题的工业基准，其标准答案通过执行经过验证的SQL程序生成）上，GRA比全上下文智能体高出5.1个百分点（准确率88.4% vs. 83.3%），同时读取的输入令牌不到后者的三分之一。无图控制实验表明，该增益主要来自智能体的选择性访问而非图拓扑，且效果取决于能够可靠驱动工具的模型。少即是多：在结构化载体上的选择性导航优于详尽的上下文。
英文摘要
Tool-calling LLM agents navigate unfamiliar codebases with a handful of generic primitives for listing, reading and searching files (ls, cat, grep). A knowledge graph admits the same interface: listing neighbours, reading node content and searching descriptions are the same operations on a different substrate. Building on this correspondence, we present GRA, a Graph Reasoning Agent that explores hybrid knowledge graphs, whose nodes are either textual concepts or relational tables, with seven generic tools, discovering everything domain-specific at run time. On UFK-M (Unified Factory Knowledge Model), an industrial benchmark of 258 analytical questions whose gold answers are produced by executing validated SQL programs, GRA beats a full-context agent by 5.1 pp (88.4% vs. 83.3%), while reading under a third of its input tokens. A graph-free control shows the gain comes chiefly from selective agentic access rather than graph topology, and that the effect depends on a model able to drive tools reliably. Seeing less, the agent answers better: selective navigation over a structured substrate beats exhaustive context.
