# SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs

arXiv: https://arxiv.org/abs/2507.07610 · HTML: https://mlsanity.com/paper/2507.07610/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 10 Jul 2025.

## Built on by (most cited first, 12 of 12)

- [Are Video Models Ready as Zero-Shot Reasoners? An Empirical Study with the MME-CoF Benchmark](https://mlsanity.com/paper/2510.26802/) (2025)
- [SpatialEvo: Self-Evolving Spatial Intelligence via Deterministic Geometric Environments](https://mlsanity.com/paper/2604.14144/) (2026)
- [How Far Are Video Models from True Multimodal Reasoning?](https://mlsanity.com/paper/2604.19193/) (2026)
- [Limits of Spatial Imagery Reasoning in Frontier LLM Models](https://mlsanity.com/paper/2603.26779/) (2026)
- [Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models](https://mlsanity.com/paper/2606.03988/) (2026)
- [Trace: A Taxonomy-Guided Environment for Multidomain Visual Reasoning](https://mlsanity.com/paper/2607.19790/) (2026)
- [Semantic Generative Tuning for Unified Multimodal Models](https://mlsanity.com/paper/2605.18714/) (2026)
- [StemBind: When MLLMs Get Lost Between Rules and Instances in Abstract Visual Reasoning](https://mlsanity.com/paper/2606.00148/) (2026)
- [MentalThink: Shaping Thoughts in Mental SVG World](https://mlsanity.com/paper/2607.03530/) (2026)
- [Does Learning Protein Folding Generalize to Broader Reasoning?](https://mlsanity.com/paper/2609.38879/) (2026)
- [WOVEN: Weaving Visual World Modeling into Multimodal LLMs](https://mlsanity.com/paper/2610.12417/) (2026)
- [S-Space: Exploring Spatial Workspace in Multimodal Language Models](https://mlsanity.com/paper/2610.13527/) (2026)

## Abstract (arXiv)

Humans can imagine and manipulate visual images mentally, a capability known as spatial visualization. While many multi-modal benchmarks assess reasoning on visible visual information, the ability to infer unseen relationships through spatial visualization remains insufficiently evaluated as a spatial skill. This reliance on publicly sourced problems from IQ tests or math competitions risks data contamination and compromises assessment reliability. To this end, we introduce SpatialViz-Bench, a comprehensive multi-modal benchmark for spatial visualization with 12 tasks across 4 sub-abilities, comprising 1,180 programmatically generated problems, a scalable framework that allows for expansion to ensure fair and continuously reliable evaluations. Our evaluation of 27 Multi-modal Large Language Models (MLLMs) reveals wide performance variations, demonstrates the benchmark's strong discriminative power, and uncovers counter-intuitive findings: Chain-of-Thought (CoT) prompting paradoxically degrades accuracy on open-source models. Through statistical and qualitative analysis of error types, SpatialViz-Bench demonstrates that state-of-the-art MLLMs exhibit deficiencies in spatial visualization tasks, thereby addressing a significant lacuna in the field. The benchmark data and evaluation code are publicly available.