Knowledge Visualization: A Benchmark and Method for Knowledge-Intensive Text-to-Image Generation

2026-04-24

AI Summary

This study addresses the challenge that existing text-to-image generation models struggle to simultaneously ensure scientific accuracy and visual plausibility in knowledge-intensive scenarios, often violating domain-specific knowledge, structural constraints, and symbolic conventions. The work presents the first systematic formulation and evaluation of this task, introducing KVBench—a knowledge visualization benchmark comprising 1,800 expert-annotated prompts spanning six high school disciplines. To enhance scientific fidelity, the authors propose KE-Check, a two-stage framework that refines prompt semantics through knowledge enrichment and guides image editing via a structured checklist. Experiments demonstrate that KE-Check effectively mitigates scientific hallucinations and significantly outperforms baseline methods in logical reasoning, symbolic precision, and multilingual robustness, while notably narrowing the performance gap between open-source and closed-source models.

Technology Category

• Computer Vision: Visual Reasoning & Symbolic Representations
• Knowledge Representation and Reasoning: Knowledge Engineering
• Data Mining & Knowledge Management: Mining of Visual, Multimedia & Multimodal Data

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics

Abstract

Recent text-to-image (T2I) models have demonstrated impressive capabilities in photorealistic synthesis and instruction following. However, their reliability in knowledge-intensive settings remains largely unexplored. Unlike natural image generation, knowledge visualization requires not only semantic alignment but also strict adherence to domain knowledge, structural constraints, and symbolic conventions, exposing a critical gap between visual plausibility and scientific correctness. To systematically study this problem, we introduce KVBench, a curriculum-grounded benchmark for evaluating knowledge-intensive T2I generation. KVBench covers six senior high-school subjects: Biology, Chemistry, Geography, History, Mathematics, and Physics. The benchmark consists of 1,800 expert-curated prompts derived from over 30 authoritative textbooks. Using this benchmark, we evaluate 14 state-of-the-art open- and closed-source models, revealing substantial deficiencies in logical reasoning, symbolic precision, and multilingual robustness, with open-source models consistently underperforming proprietary systems. To address these limitations, we further propose KE-Check, a two-stage framework that improves scientific fidelity via (1) Knowledge Elaboration for structured prompt enrichment, and (2) Checklist-Guided Refinement for explicit constraint enforcement through violation identification and constraint-guided editing. KE-Check effectively mitigates scientific hallucinations, narrowing the performance gap between open-source and leading closed-source models. Data and codes are publicly available at https://github.com/zhaoran66/KVBench.

Problem

• knowledge-intensive
• text-to-image generation
• scientific correctness
• knowledge visualization
• symbolic precision

Innovation

• knowledge visualization
• text-to-image generation
• KVBench
• KE-Check
• scientific fidelity