Coordinates from Context: Using LLMs to Ground Complex Location References
Not on the map. Why?
Abstract (arXiv)
Geocoding is the task of linking a location reference to an actual geographic location and is essential for many downstream analyses of unstructured text. In this paper, we explore the challenging setting of geocoding compositional location references. Building on recent work demonstrating LLMs' abilities to reason over geospatial data, we evaluate LLMs' geospatial knowledge versus reasoning skills relevant to our task. Based on these insights, we propose an LLM-based strategy for geocoding compositional location references. We show that our approach improves performance for the task and that a relatively small fine-tuned LLM can achieve comparable performance with much larger off-the-shelf models.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 24 of its 45 references that are in the map. First posted on arXiv 9 Oct 2025.
Builds on
24 of 45 references.
References · 24
- 2024The Llama 3 Herd of Modelscited by 14,715
- 2021LoRA: Low-Rank Adaptation of Large Language Modelscited by 14,380
- 2025Qwen3 Technical Reportcited by 8,820
- 2023Large Language Models Can Be Easily Distracted by Irrelevant Contextcited by 841
- 2025OctoThinker: Mid-training Incentivizes Reinforcement Learning Scalingcited by 113
- 2025DeepSeek-R1 Thoughtology: Let's think about LLM Reasoningcited by 92
- 2023This Land is {Your, My} Land: Evaluating Geopolitical Biases in Language Modelscited by 32
- 2023Geo-knowledge-guided GPT models improve the extraction of location descriptions from disaster-related social media messagescited by 31
- 2023Are Large Language Models Geospatially Knowledgeable?cited by 30
- 2022Geographic and Geopolitical Biases of Language Modelscited by 29
- 2025Modeling and Analyzing Urban Networks and Amenities with OSMnxcited by 25
- 2025Foundation Models for Geospatial Reasoning: Assessing Capabilities of Large Language Models in Understanding Geometries and Topological Spatial Relationscited by 23
12 more
- 2022Location reference recognition from texts: A survey and comparisoncited by 17
- 2024Geolocation Representation from Large Language Models are Generic Enhancers for Spatio-Temporal Learningcited by 15
- 2019Location reference identification from tweets during emergencies: A deep learning approachcited by 14
- 2024On the Scaling Laws of Geographical Representation in Language Modelscited by 14
- 2024Evaluating Large Language Models on Spatial Tasks: A Multi-Task Benchmarking Studycited by 12
- 2020Enhancing spatial and textual analysis with EUPEG: an extensible and unified platform for evaluating geoparserscited by 9
- 2021Changes in Twitter geolocations: Insights and suggestions for future usagecited by 7
- 2023More than Correlation: Do Large Language Models Learn Causal Representations of Space?cited by 7
- 2024Where on Earth Do Users Say They Are?: Geo-Entity Linking for Noisy Multilingual User Inputcited by 4
- 2024Quantifying Geospatial in the Common Crawl Corpuscited by 4
- 2024Privacy risk in GeoData: A surveycited by 2
- 2019Towards Geocoding Spatial Expressionscited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Understanding the Geospatial Reasoning Capabilities of LLMs: A Trajectory Recovery Perspective
- 2026GPSBench: Do Large Language Models Understand GPS Coordinates?
- 2026Network-based Spatial Context Retrieval for Open-weight LLMs: A Faithfulness Benchmark for Grounded Geographic Reasoning
- 2025Geospatial Mechanistic Interpretability of Large Language Models
- 2024GeoReasoner: Reasoning On Geospatially Grounded Context For Natural Language Understanding
- 2025Geospatial Representation Learning: A Survey from Deep Learning to The LLM Era
- 2026ReaGeo: Reasoning-Enhanced End-to-End Geocoding with LLMs
- 2025Can Large Language Models Integrate Spatial Data? Empirical Insights into Reasoning Strengths and Computational Weaknesses