# GeoChain: Multimodal Chain-of-Thought for Geographic Reasoning

arXiv: https://arxiv.org/abs/2506.00785 · HTML: https://mlsanity.com/paper/2506.00785/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 14 of its 33 references that are in the map.
- First posted on arXiv 1 Jun 2025.

## Placed in

- LLM reasoning / [Spatial and embodied reasoning](https://mlsanity.com/lr/spatial-and-embodied-reasoning/)
- Multimodal LLMs / vision-language models / [Embodied and 3D vision-language models](https://mlsanity.com/mllm/embodied-and-3d-vision-language-models/)

## Builds on (most cited first, 14 of 14)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution](https://mlsanity.com/paper/2409.12191/) (2024)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Humanity's Last Exam](https://mlsanity.com/paper/2501.14249/) (2025)
- [GEO-Bench: Toward Foundation Models for Earth Monitoring](https://mlsanity.com/paper/2306.03831/) (2023)
- [Geolocation with Real Human Gameplay Data: A Large-Scale Dataset and Human-Like Reasoning Framework](https://mlsanity.com/paper/2502.13759/) (2025)
- [AI Sees Your Location, But With A Bias Toward The Wealthy World](https://mlsanity.com/paper/2502.11163/) (2025)
- [Evaluating Precise Geolocation Inference Capabilities of Vision Language Models](https://mlsanity.com/paper/2502.14412/) (2025)
- [MAEA: Multimodal Attribution for Embodied AI](https://mlsanity.com/paper/2307.13850/) (2023)
- [HueManity: Probing Fine-Grained Visual Perception in MLLMs](https://mlsanity.com/paper/2506.03194/) (2025)
- [Semantic Augmentation in Images using Language](https://mlsanity.com/paper/2404.02353/) (2024)
- [Attribution Regularization for Multimodal Paradigms](https://mlsanity.com/paper/2404.02359/) (2024)
- [GAEA: A Geolocation Aware Conversational Assistant](https://mlsanity.com/paper/2503.16423/) (2025)

## Built on by (most cited first, 16 of 16)

- [CountQA: How Well Do MLLMs Count in the Wild?](https://mlsanity.com/paper/2508.06585/) (2025)
- [Towards Faithful Reasoning in Remote Sensing: A Perceptually-Grounded GeoSpatial Chain-of-Thought for Vision-Language Models](https://mlsanity.com/paper/2509.22221/) (2025)
- [GeoArena: Evaluating Open-World Geographic Reasoning in Large Vision-Language Models](https://mlsanity.com/paper/2509.04334/) (2025)
- [MaRVL-QA: A Benchmark for Mathematical Reasoning over Visual Landscapes](https://mlsanity.com/paper/2508.17180/) (2025)
- [The Artificial Intelligence Cognitive Examination: A Survey on the Evolution of Multimodal Evaluation from Recognition to Reasoning](https://mlsanity.com/paper/2510.04141/) (2025)
- [Explicit Reasoning Makes Better Judges: A Systematic Study on Accuracy, Efficiency, and Robustness](https://mlsanity.com/paper/2509.13332/) (2025)
- [Unlocking Zero-Shot Geospatial Reasoning via Indirect Rewards](https://mlsanity.com/paper/2510.00072/) (2025)
- [GEO-Detective: Unveiling Location Privacy Risks in Images with LLM Agents](https://mlsanity.com/paper/2511.22441/) (2025)
- [Uni-RS: A Spatially Faithful Unified Understanding and Generation Model for Remote Sensing](https://mlsanity.com/paper/2601.17673/) (2026)
- [LocationAgent: A Hierarchical Agent for Image Geolocation via Decoupling Strategy and Evidence from Parametric Knowledge](https://mlsanity.com/paper/2601.19155/) (2026)
- [UrbanGraphEmbeddings: Learning and Evaluating Spatially Grounded Multimodal Embeddings for Urban Science](https://mlsanity.com/paper/2602.08342/) (2026)
- [ENC-Bench: A Benchmark for Evaluating Multimodal Large Language Models in Electronic Navigational Chart Understanding](https://mlsanity.com/paper/2603.22763/) (2026)
- [Confidence Aware SSD Ensemble with Weighted Boxes Fusion for Weapon Detection](https://mlsanity.com/paper/2509.23697/) (2025)
- [Understanding Virality: A Rubric based Vision-Language Model Framework for Short-Form Edutainment Evaluation](https://mlsanity.com/paper/2512.21402/) (2025)
- [GTPred: Benchmarking MLLMs for Interpretable Geo-localization and Time-of-capture Prediction](https://mlsanity.com/paper/2601.13207/) (2026)
- [GeoRouter: Dynamic Paradigm Routing for Worldwide Image Geolocalization](https://mlsanity.com/paper/2603.24376/) (2026)

## Abstract (arXiv)

This paper introduces GeoChain, a large-scale benchmark for evaluating step-by-step geographic reasoning in multimodal large language models (MLLMs). Leveraging 1.46 million Mapillary street-level images, GeoChain pairs each image with a 21-step chain-of-thought (CoT) question sequence (over 30 million Q&A pairs). These sequences guide models from coarse attributes to fine-grained localization across four reasoning categories - visual, spatial, cultural, and precise geolocation - annotated by difficulty. Images are also enriched with semantic segmentation (150 classes) and a visual locatability score. Our benchmarking of contemporary MLLMs (GPT-4.1 variants, Claude 3.7, Gemini 2.5 variants) on a diverse 2,088-image subset reveals consistent challenges: models frequently exhibit weaknesses in visual grounding, display erratic reasoning, and struggle to achieve accurate localization, especially as the reasoning complexity escalates. GeoChain offers a robust diagnostic methodology, critical for fostering significant advancements in complex geographic reasoning within MLLMs.