Search papers, labs, and topics across Lattice.
This paper introduces GRA, a Graph Reasoning Agent that utilizes a schema-agnostic approach to navigate hybrid knowledge graphs composed of textual concepts and relational tables. By employing seven generic tools for operations like listing, reading, and searching, GRA effectively discovers domain-specific information at runtime. On the UFK-M benchmark, GRA outperformed a full-context agent by 5.1 percentage points while processing significantly fewer input tokens, demonstrating that selective access to information can enhance reasoning accuracy.
Selective navigation in hybrid knowledge graphs allows agents to outperform full-context models, achieving better results with less information.
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.
