When Vision Meets Graphs: A Survey on Graph Reasoning and Learning
cs.SI, cs.CV, cs.LG
Submitted: 2026-09-03
Updated: 2026-09-03
Comments: IJCAI Survey Track, 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Graphs are a fundamental data structure underlying many problems in the natural and social sciences.
Terminology
Abstract
Graphs are a fundamental data structure underlying many problems in the natural and social sciences. Over the past decade, Graph Neural Networks (GNNs) have dominated graph machine learning, supported by solid theoretical foundations. Yet scientists often understand graph structure through vision: chemists read molecular diagrams and social scientists inspect network visualizations. Despite decades of work on graph visualization, most graph learning pipelines still treat graphs purely as symbolic structures, rarely leveraging the visual form of graphs. We argue that this gap deserves renewed attention in the era of powerful vision and vision-language models. This survey provides a first systematic overview of the emerging area we term vision meets graphs, which treats visual depictions of graphs as first-class inputs for reasoning and learning. We organize existing work into three threads. Vision for Graph Reasoning studies how models can use visual depictions of graphs to understand structure and carry out multi-step reasoning. Vision for Graph Learning explores how visual features can complement or augment graph encoders beyond known limitations of message passing. Scientific Graphs examines domains where standardized depiction conventions support both reasoning and learning. Our goal is to clarify what current methods can and cannot do, and to outline a path toward foundation models that perceive and reason about graphs as scientists do.
Sources
- MolVision: Molecular Property Prediction with Vision Language Models
- AI-Driven Automation Can Become the Foundation of Next-Era Science of Science Research
- Chemception: A Deep Neural Network with Minimal Chemistry Knowledge Matches the Performance of Expert-developed QSAR/QSPR Models
- See or Say Graphs: Agent-Driven Scalable Graph Structure Understanding with Vision-Language Models
- Lost in Serialization: Invariance and Generalization of LLM Graph Reasoners
- Rethinking Spectral Augmentation for Contrast-based Graph Self-Supervised Learning
- Graph-to-Vision: Multi-graph Understanding and Reasoning using Vision-Language Models
- Spectral Augmentation for Self-Supervised Learning on Graphs
- GraphInstruct: Empowering Large Language Models with Graph Understanding and Reasoning Capability
- Chain-of-Visual-Thought: Teaching VLMs to See and Think Better with Continuous Visual Tokens
- LiveProteinBench: A Contamination-Free Benchmark for Assessing Models' Specialized Capabilities in Protein Science
- VisGraphVar: A Benchmark Generator for Assessing Variability in Graph Analysis Using Large Vision-Language Models
- GraphBench: Next-generation graph learning benchmarking
- Thinking with Images for Multimodal Reasoning: Foundations, Methods, and Future Frontiers
- GraphVerse: A Comprehensive Visual Graph Reasoning Benchmark for Multimodal Large Language Models
- Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
- Multimodal LLMs Struggle with Basic Visual Network Analysis: a VNA Benchmark
- How Powerful are Graph Neural Networks?
- GraphOmni: A Comprehensive and Extensible Benchmark Framework for Large Language Models on Graph-theoretic Tasks
- Scope: Selective Cross-modal Orchestration of Visual Perception Experts
Related papers
- Linking Scalar-Intensity Language to Structural Polarization with Validated Signed-Network Measures
- Detection and Characterization of Coordinated Online Behavior: A Survey
- Omega-N: Interpretable Structural Node Descriptors and Their Applicability Domain
- Simplify to Amplify: Achieving Information-Theoretic Bounds with Fewer Steps in Spectral Community Detection
- Transmission Neural Networks: Inhibitory and Excitatory Connections
- A family of graph GOSPA metrics for graphs with different sizes