SolarGPT-QA: A Domain-Adaptive Large Language Model for Educational Question Answering in Space Weather and Heliophysics

TL;DR

SolarGPT-QA tackles educational question answering in space weather and heliophysics by domain-adapting a large language model with LoRA and fine-tuning on pedagogically structured QA data. It leverages a curated heliophysics literature corpus (2010–2024) and synthetic QA generation to produce age-appropriate explanations while maintaining scientific grounding. Human evaluations show SolarGPT-QA outperforms zero-shot general-purpose models and achieves competitive performance relative to instruction-tuned baselines, with ablations demonstrating the value of combining domain-adaptive pretraining and educational fine-tuning. The work outlines a roadmap toward SolarGPT's broader educational and forecasting ecosystem, including core, multimodal, and retrieval-enhanced extensions.

Abstract

Solar activity, including solar flares, coronal mass ejections (CMEs), and geomagnetic storms, can significantly impact satellites, aviation, power grids, data centers, and space missions. Extreme solar events can cause substantial economic damage if not predicted in advance, highlighting the importance of accurate forecasting and effective education in space science. Although large language models (LLMs) perform well on general tasks, they often lack domain-specific knowledge and pedagogical capability to clearly explain complex space science concepts. We introduce SolarGPT-QA, a question answering system based on a domain-adapted large language model built on the LLaMA-3 base model. The model is trained using scientific literature and large-scale question-answer data generated with GPT-4 and refined using Grok-3 in a student-friendly storytelling style. Human pairwise evaluations show that SolarGPT-QA outperforms general-purpose models in zero-shot settings and achieves competitive performance compared to instruction-tuned models for educational explanations in space weather and heliophysics. A small pilot student comprehension study further suggests improved clarity and accessibility of the generated explanations. Ablation experiments indicate that combining domain-adaptive pretraining with pedagogical fine-tuning is important for balancing scientific accuracy and educational effectiveness. This work represents an initial step toward a broader SolarGPT framework for space science education and forecasting.

Figures (6)

• Figure 1: Overview of the SolarGPT-QA system illustrating its educational question–answering framework for space weather and heliophysics.
• Figure 2: System architecture and workflow of SolarGPT-QA, showing the data sources, domain adaptation, and question–answer generation pipeline.
• Figure 3: Combination of Eruption Event Types as Paper Keywords
• Figure 4: WordCloud of the data corpus used for DAPT
• Figure 5: Human evaluation win-rate comparison across models.