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

TL;DR

This paper introduces KVBench, a comprehensive benchmark for evaluating knowledge-intensive text-to-image models across science subjects, and proposes KE-Check, a framework to enhance scientific accuracy and reduce hallucinations in generated images.

Contribution

The paper presents KVBench, a new benchmark with expert-curated prompts for knowledge-intensive T2I tasks, and KE-Check, a novel two-stage framework to improve scientific fidelity in image generation.

Findings

Open-source models underperform proprietary ones in knowledge-intensive tasks.

KE-Check reduces scientific hallucinations and improves model fidelity.

Benchmark reveals deficiencies in logical reasoning and multilingual robustness.

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.