# Do Multimodal Language Models Really Understand Direction? A Benchmark for Compass Direction Reasoning

arXiv: https://arxiv.org/abs/2412.16599 · HTML: https://mlsanity.com/paper/2412.16599/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning.
- Builds on 7 of its 27 references that are in the map.
- First posted on arXiv 21 Dec 2024.

## 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, 7 of 7)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [World Models](https://mlsanity.com/paper/1803.10122/) (2018)
- [Reasoning with Language Model is Planning with World Model](https://mlsanity.com/paper/2305.14992/) (2023)
- [What's "up" with vision-language models? Investigating their struggle with spatial reasoning](https://mlsanity.com/paper/2310.19785/) (2023)
- [World Model on Million-Length Video And Language With Blockwise RingAttention](https://mlsanity.com/paper/2402.08268/) (2024)
- [SpartQA: : A Textual Question Answering Benchmark for Spatial Reasoning](https://mlsanity.com/paper/2104.05832/) (2021)

## Built on by (most cited first, 3 of 3)

- [AVA-Bench: Atomic Visual Ability Benchmark for Vision Foundation Models](https://mlsanity.com/paper/2506.09082/) (2025)
- [Evaluating the Ability of Large Language Models to Reason about Cardinal Directions, Revisited](https://mlsanity.com/paper/2507.12059/) (2025)
- [Stitch and Tell: A Structured Multimodal Data Augmentation Method for Spatial Understanding](https://mlsanity.com/paper/2512.06769/) (2025)

## Abstract (arXiv)

Direction reasoning is essential for intelligent systems to understand the real world. While existing work focuses primarily on spatial reasoning, compass direction reasoning remains underexplored. To address this, we propose the Compass Direction Reasoning (CDR) benchmark, designed to evaluate the direction reasoning capabilities of multimodal language models (MLMs). CDR includes three types images to test spatial (up, down, left, right) and compass (north, south, east, west) directions. Our evaluation reveals that most MLMs struggle with direction reasoning, often performing at random guessing levels. Experiments show that training directly with CDR data yields limited improvements, as it requires an understanding of real-world physical rules. We explore the impact of mixdata and CoT fine-tuning methods, which significantly enhance MLM performance in compass direction reasoning by incorporating diverse data and step-by-step reasoning, improving the model's ability to understand direction relationships.