SVG Decomposition for Enhancing Large Multimodal Models Visualization Comprehension: A Study with Floor Plans
On the map: LLM reasoning → Spatial and embodied reasoning
Abstract (arXiv)
Large multimodal models (LMMs) are increasingly capable of interpreting visualizations, yet they continue to struggle with spatial reasoning. One proposed strategy is decomposition, which breaks down complex visualizations into structured components. In this work, we examine the efficacy of scalable vector graphics (SVGs) as a decomposition strategy for improving LMMs' performance on floor plans comprehension. Floor plans serve as a valuable testbed because they combine geometry, topology, and semantics, and their reliable comprehension has real-world applications, such as accessibility for blind and low-vision individuals. We conducted an exploratory study with three LMMs (GPT-4o, Claude 3.7 Sonnet, and Llama 3.2 11B Vision Instruct) across 75 floor plans. Results show that combining SVG with raster input (SVG+PNG) improves performance on spatial understanding tasks but often hinders spatial reasoning, particularly in pathfinding. These findings highlight both the promise and limitations of decomposition as a strategy for advancing spatial visualization comprehension.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Spatial and embodied reasoning, one of 17 directions in LLM reasoning. Builds on 17 of its 36 references that are in the map. First posted on arXiv 5 Nov 2025.
Builds on
17 of 36 references.
Closest ancestors · 3
Same direction, newest first
- 2024TopViewRS: Vision-Language Models as Top-View Spatial Reasoners4 Jun
- 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilities22 Jan
- 2023What's "up" with vision-language models? Investigating their struggle with spatial reasoning30 Oct
Other references · 14
- 2022ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoningcited by 1,449
- 2019CubiCasa5K: A Dataset and an Improved Multi-Task Model for Floorplan Image Analysiscited by 49
- 2025Can Large Vision Language Models Read Maps Like a Human?cited by 30
- 2019Deep Floor Plan Recognition Using a Multi-Task Network with Room-Boundary-Guided Attentioncited by 27
- 2024Advancing Multimodal Large Language Models in Chart Question Answering with Visualization-Referenced Instruction Tuningcited by 27
- 2023How Good is ChatGPT in Giving Advice on Your Visualization Design?cited by 20
- 2024Chart-based Reasoning: Transferring Capabilities from LLMs to VLMscited by 20
- 2024Exploring the Capability of LLMs in Performing Low-Level Visual Analytic Tasks on SVG Data Visualizationscited by 18
- 2024ImplicitAVE: An Open-Source Dataset and Multimodal LLMs Benchmark for Implicit Attribute Value Extractioncited by 17
- 2023Leveraging Large Language Models for Scalable Vector Graphics-Driven Image Understandingcited by 10
- 2024FloNa: Floor Plan Guided Embodied Visual Navigationcited by 10
- 2025Eliminating Rasterization: Direct Vector Floor Plan Generation with DiffPlannercited by 4
2 more
- 2025Research on Navigation Methods Based on LLMscited by 3
- 2021Knowledge driven Description Synthesis for Floor Plan Interpretationcited by 2
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2024Visually Descriptive Language Model for Vector Graphics Reasoning
- 2025SpatialViz-Bench: A Cognitively-Grounded Benchmark for Diagnosing Spatial Visualization in MLLMs
- 2024VGBench: Evaluating Large Language Models on Vector Graphics Understanding and Generation
- 2026SpaceDG: Benchmarking Spatial Intelligence under Visual Degradation
- 2026MentalBlackboard: Evaluating Spatial Visualization via Mathematical Transformations
- 2025Jigsaw-Puzzles: From Seeing to Understanding to Reasoning in Vision-Language Models
- 2025Unfolding Spatial Cognition: Evaluating Multimodal Models on Visual Simulations
- 2025Multimodal Spatial Reasoning in the Large Model Era: A Survey and Benchmarks