# Foundation Models for Geospatial Reasoning: Assessing Capabilities of Large Language Models in Understanding Geometries and Topological Spatial Relations

arXiv: https://arxiv.org/abs/2505.17136 · HTML: https://mlsanity.com/paper/2505.17136/

## Key facts

- Cited by 23 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 41 of its 113 references that are in the map.
- First posted on arXiv 22 May 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)

## Builds on (most cited first, 30 of 41)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Flamingo: a Visual Language Model for Few-Shot Learning](https://mlsanity.com/paper/2204.14198/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [LaMDA: Language Models for Dialog Applications](https://mlsanity.com/paper/2201.08239/) (2022)
- [Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces](https://mlsanity.com/paper/2412.14171/) (2024)
- [Text and Code Embeddings by Contrastive Pre-Training](https://mlsanity.com/paper/2201.10005/) (2022)
- [GatorTron: A Large Clinical Language Model to Unlock Patient Information from Unstructured Electronic Health Records](https://mlsanity.com/paper/2203.03540/) (2022)
- [SGPT: GPT Sentence Embeddings for Semantic Search](https://mlsanity.com/paper/2202.08904/) (2022)
- [An efficient framework for learning sentence representations](https://mlsanity.com/paper/1803.02893/) (2018)
- [Query2doc: Query Expansion with Large Language Models](https://mlsanity.com/paper/2303.07678/) (2023)
- [Foundation Models for Generalist Geospatial Artificial Intelligence](https://mlsanity.com/paper/2310.18660/) (2023)
- [Evaluating Spatial Understanding of Large Language Models](https://mlsanity.com/paper/2310.14540/) (2023)
- [GeoLLM: Extracting Geospatial Knowledge from Large Language Models](https://mlsanity.com/paper/2310.06213/) (2023)
- [A Review of Location Encoding for GeoAI: Methods and Applications](https://mlsanity.com/paper/2111.04006/) (2021)
- [Autonomous GIS: the next-generation AI-powered GIS](https://mlsanity.com/paper/2305.06453/) (2023)
- [Adversarial Text-to-Image Synthesis: A Review](https://mlsanity.com/paper/2101.09983/) (2021)
- [Geo-knowledge-guided GPT models improve the extraction of location descriptions from disaster-related social media messages](https://mlsanity.com/paper/2310.09340/) (2023)
- [Towards Reproducible LLM Evaluation: Quantifying Uncertainty in LLM Benchmark Scores](https://mlsanity.com/paper/2410.03492/) (2024)
- [SE-KGE: A Location-Aware Knowledge Graph Embedding Model for Geographic Question Answering and Spatial Semantic Lifting](https://mlsanity.com/paper/2004.14171/) (2020)
- [Dialectical language model evaluation: An initial appraisal of the commonsense spatial reasoning abilities of LLMs](https://mlsanity.com/paper/2304.11164/) (2023)
- [Assessment of a new GeoAI foundation model for flood inundation mapping](https://mlsanity.com/paper/2309.14500/) (2023)

## Built on by (most cited first, 23 of 23)

- [GeoAnalystBench: A GeoAI benchmark for assessing large language models for spatial analysis workflow and code generation](https://mlsanity.com/paper/2509.05881/) (2025)
- [GeoAgentBench: A Dynamic Execution Benchmark for Tool-Augmented Agents in Spatial Analysis](https://mlsanity.com/paper/2604.13888/) (2026)
- [A Vision for Geo-Temporal Deep Research Systems: Towards Comprehensive, Transparent, and Reproducible Geo-Temporal Information Synthesis](https://mlsanity.com/paper/2506.14345/) (2025)
- [GeoResponder: Towards Building Geospatial LLMs for Time-Critical Disaster Response](https://mlsanity.com/paper/2509.19354/) (2025)
- [Georeferencing complex relative locality descriptions with large language models](https://mlsanity.com/paper/2512.14228/) (2025)
- [Thinking on Maps: How Foundation Model Agents Explore, Remember, and Reason Map Environments](https://mlsanity.com/paper/2512.24504/) (2025)
- [Mil-SCORE: Benchmarking Long-Context Geospatial Reasoning and Planning in Large Language Models](https://mlsanity.com/paper/2601.21826/) (2026)
- [A Dual-Helix Governance Approach Towards Reliable Agentic Artificial Intelligence for WebGIS Development](https://mlsanity.com/paper/2603.04390/) (2026)
- [Unbox Responsible GeoAI: Navigating Climate Extreme and Disaster Mapping](https://mlsanity.com/paper/2605.00315/) (2026)
- [Spatial Representation Learning Beyond Pixels: Unifying Raster Data and Vector Semantics for Human-Centric Geospatial Foundation Models](https://mlsanity.com/paper/2606.02374/) (2026)
- [From Pixels to Urban Policy-Intelligence: Recovering Legacy Effects of Redlining with a Multimodal LLM](https://mlsanity.com/paper/2509.15132/) (2025)
- [Coordinates from Context: Using LLMs to Ground Complex Location References](https://mlsanity.com/paper/2510.08741/) (2025)
- [Geography According to ChatGPT -- How Generative AI Represents and Reasons about Geography](https://mlsanity.com/paper/2603.18881/) (2026)
- [Assessing the Geographic Diversity of AI's Platial Representations in Image Generation](https://mlsanity.com/paper/2606.05188/) (2026)
- [Multimodal and Multiscale Spatial-Temporal Semantic Search and Recommendation with AI Foundation Models](https://mlsanity.com/paper/2606.28369/) (2026)
- [Do Maps Still Matter for Machines: Revisiting the Role of Choropleth Maps in Foundation Model Spatial Understanding](https://mlsanity.com/paper/2607.17999/) (2026)
- [LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents](https://mlsanity.com/paper/2608.06948/) (2026)
- [Geo-Spatial Concept Probing of Large Language Models: Abstraction, Compositionality, and Grounding](https://mlsanity.com/paper/2608.07353/) (2026)
- [Physico-Geospatial Grounded Scene Interpretation for Mobile Robotics](https://mlsanity.com/paper/2609.06629/) (2026)
- [PolyTopoBench: A Benchmark for Complex Vector Polygon Generation from Remote Sensing Imagery](https://mlsanity.com/paper/2609.32856/) (2026)
- [Mind the Spatial Gap: A Benchmark for Geospatial Reasoning and Constrained Spatial Editing](https://mlsanity.com/paper/2610.12881/) (2026)
- [A Decade of PhotoMappers: A Longitudinal Study of Crowdsourced Disaster Photos Geolocalization with Vision-Language Models and Geospatial Reasoning](https://mlsanity.com/paper/2610.15403/) (2026)
- [Geospatial Vector Geometry as a Native Spatial Modality for Vision-Language Models](https://mlsanity.com/paper/2610.16769/) (2026)

## Abstract (arXiv)

Applying AI foundation models directly to geospatial datasets remains challenging due to their limited ability to represent and reason with geographical entities, specifically vector-based geometries and natural language descriptions of complex spatial relations. To address these issues, we investigate the extent to which a well-known-text (WKT) representation of geometries and their spatial relations (e.g., topological predicates) are preserved during spatial reasoning when the geospatial vector data are passed to large language models (LLMs) including GPT-3.5-turbo, GPT-4, and DeepSeek-R1-14B. Our workflow employs three distinct approaches to complete the spatial reasoning tasks for comparison, i.e., geometry embedding-based, prompt engineering-based, and everyday language-based evaluation. Our experiment results demonstrate that both the embedding-based and prompt engineering-based approaches to geospatial question-answering tasks with GPT models can achieve an accuracy of over 0.6 on average for the identification of topological spatial relations between two geometries. Among the evaluated models, GPT-4 with few-shot prompting achieved the highest performance with over 0.66 accuracy on topological spatial relation inference. Additionally, GPT-based reasoner is capable of properly comprehending inverse topological spatial relations and including an LLM-generated geometry can enhance the effectiveness for geographic entity retrieval. GPT-4 also exhibits the ability to translate certain vernacular descriptions about places into formal topological relations, and adding the geometry-type or place-type context in prompts may improve inference accuracy, but it varies by instance. The performance of these spatial reasoning tasks offers valuable insights for the refinement of LLMs with geographical knowledge towards the development of geo-foundation models capable of geospatial reasoning.