Graph-Enhanced Large Language Models for Spatial Search
2026-06-22 • Databases
DatabasesArtificial IntelligenceInformation Retrieval
AI summaryⓘ
The authors explain that while large language models (LLMs) are getting better at answering complex questions, they still struggle with spatial reasoning, which means understanding and reasoning about physical spaces. This skill is important for areas like urban planning and engineering. They point out that spatial information is often stored as graphs and suggest new research should focus on teaching LLMs to work with these graphs. The paper imagines a future where search engines and LLMs work together to solve tough spatial problems.
Large Language ModelsSpatial ReasoningRetrieval Augmented GenerationGraph DataUrban PlanningCivil EngineeringSearch EnginesGraph-Enhanced Reasoning
Authors
Nicole R. Schneider, Kent O'Sullivan, Hanan Samet
Abstract
There have been many recent improvements in the ability of Large Language Models (LLMs) to perform complex tasks and answer domain-specific questions through techniques like Retrieval Augmented Generation (RAG). However, reasoning abilities of LLMs, including spatial reasoning abilities, are still lacking. Spatial reasoning is a key component required to answer questions in a variety of domains that are grounded in the physical world, including urban planning, civil engineering, travel, and many others. To advance the development of LLMs and facilitate an impact in these domains, new research techniques must be developed to enable LLMs to reason over spatial data, which is commonly stored in the form of a graph. In this paper we outline the challenges associated with spatial reasoning through LLMs and envision a future in which search engines integrate with LLMs to answer complex spatial questions through graph-enhanced reasoning.