Artificial intelligence (AI) is revolutionizing astronomy by enhancing data analysis and public engagement. A notable development is AstroLLaMA, a 7-billion-parameter model fine-tuned from LLaMA-2 using over 300,000 astronomy abstracts from arXiv.
AstroLLaMA achieves a 30% lower perplexity than Llama-2, showing marked domain adaptation. It generates more insightful and scientifically relevant text completions and embedding extraction than state-of-the-art foundation models, despite having significantly fewer parameters.
AstroLLaMA serves as a robust, domain-specific model with broad fine-tuning potential, aiming to spur astronomy-focused research, including automatic paper summarization and conversational agent development.
