🧠 AI Ed Wiki

📄 Research Article

NuclearDiffusion: Text-to-Image Foundation Models for Learning Nuclear Energy Concepts

● high · 2026-08-06generative-ai, stem-education, ai-generated-content, multimodal, content-quality, higher-ed, visualization

Synthesis: Systematic study of domain-adapted text-to-image models for nuclear engineering education. Fine-tunes Stable Diffusion on nuclear domain images; fine-tuned model achieves 78% domain accuracy vs 12% for base model. Proposes NuclearDiffusion as an educational tool where instructors generate accurate visualizations of nuclear concepts (reactor components, fuel cycles, safety systems). Demonstrates that domain-specific fine-tuning dramatically improves visual correctness for specialized STEM concept illustration. Generative AI, Generative AI, STEM Education, content-quality, and Multimodal.

Systematic study of domain-adapted text-to-image models for nuclear engineering education. Fine-tunes Stable Diffusion on nuclear domain images; fine-tuned model achieves 78% domain accuracy vs 12% for base model. Proposes NuclearDiffusion as an educational tool where instructors generate accurate visualizations of nuclear concepts (reactor components, fuel cycles, safety systems). Demonstrates that domain-specific fine-tuning dramatically improves visual correctness for specialized STEM concept illustration.

Connected Concepts

  • Generative AI
  • STEM Education
  • Multimodal

    • Connected Articles

  • Mllm Scientific Visualization Literacy — Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy
  • Anvil AI Educational Animations — ANVIL: Analogies and Videos for Lecturers
  • GenAI Runaway Object Math Higher Ed — GenAI as a runaway object in higher education: A socio-cultural view on AI-influenced academic practice in mathematics
  • Curriculum As Code Instructional Design 2026 — Curriculum as Code: An AI-Assisted Architecture for Instructional Design in STEM Education
  • Lata Ferpa Compliant Local LLM Autograder — LaTA: A Drop-in, FERPA-Compliant Local-LLM Autograder for Upper-Division STEM Coursework
  • Prompt Based Programming Lesson — Teaching Prompt-Based Programming with LLMs: A 45-Minute Lesson with Guided Practice for End-User Programmers

    • Citation

    Mohammed I. Radaideh, Jeremy Moon, Andre Gala-Garza, Emma Son, Yug Shah, & Majdi I. Radaideh (2026). NuclearDiffusion: Text-to-Image Foundation Models for Learning Nuclear Energy Concepts. arXiv:2608.04030. arXiv preprint (cross-listed cs.GR/cs.CY).