Large language models are increasingly used as educational assistants, yet evaluation of their educational capabilities remains concentrated on question-answering and tutoring tasks. A critical gap exists for multimedia instructional content generation -- the ability to produce coherent, diagram-rich explanations that combine geometrically accurate visuals with step-by-step reasoning. We present EduIllustrate, a benchmark for evaluating LLMs on interleaved text-diagram explanation generation for K-12 STEM problems. The benchmark comprises 230 problems spanning five subjects and three grade levels, a standardized generation protocol with sequential anchoring to enforce cross-diagram visual consistency, and an 8-dimension evaluation rubric grounded in multimedia learning theory covering both text and visual quality. Evaluation of ten LLMs reveals a wide performance spread: Gemini 3.0 Pro Preview leads at 87.8\%, while Kimi-K2.5 achieves the best cost-efficiency (80.8\% at \$0.12/problem). Workflow ablation confirms sequential anchoring improves Visual Consistency by 13\% at 94\% lower cost. Human evaluation with 20 expert raters validates LLM-as-judge reliability for objective dimensions (ρ≥0.83) while revealing limitations on subjective visual assessment.
@article{arxiv.2604.05005,
title = {EduIllustrate: Towards Scalable Automated Generation Of Multimodal Educational Content},
author = {Shuzhen Bi and Mingzi Zhang and Zhuoxuan Li and Xiaolong Wang and Keqian Li and Aimin Zhou},
journal= {arXiv preprint arXiv:2604.05005},
year = {2026}
}