# Exploring and Improving the Spatial Reasoning Abilities of Large Language Models

arXiv: https://arxiv.org/abs/2312.01054 · HTML: https://mlsanity.com/paper/2312.01054/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 0 of its 25 references that are in the map.
- First posted on arXiv 2 Dec 2023.

## Placed in

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

## Built on by (most cited first, 9 of 9)

- [Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks](https://mlsanity.com/paper/2504.18838/) (2025)
- [Evaluating Large Language Models on Spatial Tasks: A Multi-Task Benchmarking Study](https://mlsanity.com/paper/2408.14438/) (2024)
- [A Survey of Large Language Model-Powered Spatial Intelligence Across Scales: Advances in Embodied Agents, Smart Cities, and Earth Science](https://mlsanity.com/paper/2504.09848/) (2025)
- [GRASP: A Grid-Based Benchmark for Evaluating Commonsense Spatial Reasoning](https://mlsanity.com/paper/2407.01892/) (2024)
- [MosaicThinker: On-Device Visual Spatial Reasoning for Embodied AI via Iterative Construction of Space Representation](https://mlsanity.com/paper/2602.07082/) (2026)
- [From Human Cognition to Neural Activations: Probing the Computational Primitives of Spatial Reasoning in LLMs](https://mlsanity.com/paper/2603.26323/) (2026)
- [Large Language Models and their Awareness of Mechanics and Spatial Geometry](https://mlsanity.com/paper/2608.14615/) (2026)
- [Block-level Text Spotting with LLMs](https://mlsanity.com/paper/2406.13208/) (2024)
- [AlphaSpace: Enabling Robotic Actions through Semantic Tokenization and Symbolic Reasoning](https://mlsanity.com/paper/2503.18769/) (2025)

## Abstract (arXiv)

Large Language Models (LLMs) represent formidable tools for sequence modeling, boasting an innate capacity for general pattern recognition. Nevertheless, their broader spatial reasoning capabilities, especially applied to numerical trajectory data, remain insufficiently explored. In this paper, we investigate the out-of-the-box performance of ChatGPT-3.5, ChatGPT-4 and Llama 2 7B models when confronted with 3D robotic trajectory data from the CALVIN baseline and associated tasks, including 2D directional and shape labeling. Additionally, we introduce a novel prefix-based prompting mechanism, which yields a 33% improvement on the 3D trajectory data and an increase of up to 10% on SpartQA tasks over zero-shot prompting (with gains for other prompting types as well). The experimentation with 3D trajectory data offers an intriguing glimpse into the manner in which LLMs engage with numerical and spatial information, thus laying a solid foundation for the identification of target areas for future enhancements.