Spectra as Language: Large Language Models for Scalable Stellar Parameter and Abundance Inference

2026-05-21

AI Summary

Traditional approaches to analyzing large-scale stellar spectra are hindered by high-dimensional data, limited generalization, and low computational efficiency. This work introduces large language models (LLMs) into stellar spectral analysis for the first time, proposing a two-stage, scalable sequence modeling framework that treats spectra as continuous signals to enable end-to-end prediction of stellar effective temperature, surface gravity, metallicity, and abundances of approximately 20 chemical elements. By fully leveraging the sequential nature of spectral data, the method achieves high-precision estimates on large datasets, and scaling law analysis demonstrates that its performance systematically improves with increasing data volume. This study establishes an efficient and scalable new paradigm for astronomical spectral analysis.

Abstract

Stellar spectra encode key information on the physical properties and chemical compositions of stars. Accurate stellar parameter determination is essential for addressing major questions such as galaxy and stellar evolution. Large-scale spectroscopic surveys have accumulated unprecedented spectral data. Traditional feature extraction or model-fitting approaches struggle with high-dimensional, massive datasets, limited generalization, and computational inefficiency. Recent advances in large language models demonstrate strong generalization and feature-learning in tasks like natural language processing, DNA/RNA sequence analysis, and protein/chemical parsing. Stellar spectra are continuous sequential signals, enabling the transfer of language models to stellar spectroscopy. Here, we propose a two-stage large language model framework for stellar parameter inference, achieving accurate estimation of effective temperature, surface gravity, metallicity, and abundances of ~20 chemical elements. Scaling-law analyses show systematic performance improvements with increasing data, providing a scalable framework for forthcoming large-scale surveys.

Problem

• stellar spectra
• stellar parameter inference
• chemical abundances
• large-scale spectroscopic surveys
• high-dimensional data

Innovation

• large language models
• stellar spectroscopy
• stellar parameter inference
• scaling laws
• spectral analysis