ERUnderstand: Evaluating Vision-Language Models on Structured ER Diagrams
2026-07-27 • Artificial Intelligence
Artificial IntelligenceComputer Vision and Pattern RecognitionDatabases
AI summaryⓘ
The authors created ERUnderstand, a large dataset of nearly 3,000 Entity-Relationship Diagrams (ERDs) along with machine-readable versions to help test how well AI models can understand these diagrams. They tested current vision-language models and found that while these models identify common parts of ERDs fairly well, they struggle with complicated elements like weak entities and multivalued attributes. Adding reasoning helps improve results, but challenges remain, especially as diagrams get more complex. Their work provides a useful benchmark for improving AI understanding of database designs.
Entity-Relationship DiagramConceptual Database DesignVision-Language ModelsMachine-readable SchemaWeak EntitiesMultivalued AttributesN-ary RelationshipsBenchmark DatasetReasoning-Augmented Models
Authors
Ali Ansari, Yasmin Mohammadi, Farnoush Nili, Parsa Esmaeilkhani, Longin Jan Latecki, Eduard Dragut
Abstract
Entity-Relationship Diagrams (ERDs) are central to conceptual database design, yet they are typically available only as rendered images rather than machine-readable schemas, limiting AI-assisted database engineering. We introduce ERUnderstand, the first large-scale benchmark for structured understanding of ER diagrams, comprising 2,960 diagrams collected from curated educational sources, real-world schemas, and synthetically generated examples spanning diverse domains, notations, complexity levels, and Extended Entity-Relationship (EER) constructs. Each diagram is paired with a standardized machine-readable representation for fine-grained evaluation of schema elements. Evaluating state-of-the-art Vision-Language Models (VLMs), we find that while common ERD elements are recovered reliably (F1 > 0.74), performance drops sharply on weak entities (as low as 0.28 F1), multivalued attributes (0.14 F1), and N-ary relationships (0.07 F1). Reasoning-augmented models improve overall performance by 15-25% but remain sensitive to linguistic priors and increasing diagram complexity. ERUnderstand provides a standardized benchmark for evaluating multimodal understanding of conceptual database schemas. The benchmark, dataset, evaluation toolkit, and generation code are publicly available at https://github.com/salinaria/ERUnderstand.