English

Finetuning Stellar Spectra Foundation Models with LoRA

Instrumentation and Methods for Astrophysics 2025-07-29 v1 Solar and Stellar Astrophysics

Abstract

Foundation models are beginning to impact stellar spectroscopy, where spectra encode rich physical information in a structured, language-like form. A key challenge is adapting these models across heterogeneous surveys with differing resolution and coverage. We apply Low-Rank Adaptation (LoRA) to fine-tune SpecCLIP--a contrastively pre-trained model on LAMOST and Gaia XP spectra--for downstream tasks on DESI Early Data Release (EDR) spectra. We show that LoRA enables few-shot learning on DESI, with performance varying by fine-tuned module and benefiting from Gaia XP knowledge embedded in the pre-trained model. Our results demonstrate that LoRA provides a lightweight and effective strategy for extending spectral foundation models to new instruments and survey domains.

Cite

@article{arxiv.2507.20972,
  title  = {Finetuning Stellar Spectra Foundation Models with LoRA},
  author = {Xiaosheng Zhao and Yuan-Sen Ting and Alexander S. Szalay and Yang Huang},
  journal= {arXiv preprint arXiv:2507.20972},
  year   = {2025}
}

Comments

7 pages, 2 figures. Accepted to the Machine Learning for Astrophysics (ML4Astro) Colocated Workshop at ICML 2025. Presented as a spotlight talk

R2 v1 2026-07-01T04:22:22.768Z