Grid Spatial Understanding: A Dataset for Textual Spatial Reasoning over Grids, Embodied Settings, and Coordinate Structures
This work addresses the need for better spatial reasoning benchmarks in AI, particularly for embodied agents, though it is incremental as it builds on existing datasets and methods.
The authors tackled the problem of evaluating spatial reasoning in LLMs by introducing the GSU dataset, which tests navigation, object localization, and structure composition without visual inputs, finding that models struggle with frames of reference and 3D shapes, but fine-tuning small models can match frontier performance.
We introduce GSU, a text-only grid dataset to evaluate the spatial reasoning capabilities of LLMs over 3 core tasks: navigation, object localization, and structure composition. By forgoing visual inputs, isolating spatial reasoning from perception, we show that while most models grasp basic grid concepts, they struggle with frames of reference relative to an embodied agent and identifying 3D shapes from coordinate lists. We also find that exposure to a visual modality does not provide a generalizable understanding of 3D space that VLMs are able to utilize for these tasks. Finally, we show that while the very latest frontier models can solve the provided tasks (though harder variants may still stump them), fully fine-tuning a small LM or LORA fine-tuning a small LLM show potential to match frontier model performance, suggesting an avenue for specialized embodied agents.