Saved in:
Bibliographic Details
Main Authors: Zhao, Xiaosheng, Ting, Yuan-Sen, Wyse, Rosemary F. G., Szalay, Alexander S., Huang, Yang, Dobos, László, Budavári, Tamás, Wei, Viska
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.15021
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911451060371456
author Zhao, Xiaosheng
Ting, Yuan-Sen
Wyse, Rosemary F. G.
Szalay, Alexander S.
Huang, Yang
Dobos, László
Budavári, Tamás
Wei, Viska
author_facet Zhao, Xiaosheng
Ting, Yuan-Sen
Wyse, Rosemary F. G.
Szalay, Alexander S.
Huang, Yang
Dobos, László
Budavári, Tamás
Wei, Viska
contents Cross-survey generalization is a critical challenge in stellar spectral analysis, particularly in cases such as transferring from low- to moderate-resolution surveys. We investigate this problem using pre-trained models, focusing on simple neural networks such as multilayer perceptrons (MLPs), with a case study transferring from LAMOST low-resolution spectra (LRS) to DESI medium-resolution spectra (MRS). Specifically, we pre-train MLPs on either LRS or their embeddings and fine-tune them for application to DESI stellar spectra. We compare MLPs trained directly on spectra with those trained on embeddings derived from transformer-based models (self-supervised foundation models pre-trained for multiple downstream tasks). We also evaluate different fine-tuning strategies, including residual-head adapters, LoRA, and full fine-tuning. We find that MLPs pre-trained on LAMOST LRS achieve strong performance, even without fine-tuning, and that modest fine-tuning with DESI spectra further improves the results. For iron abundance, embeddings from a transformer-based model yield advantages in the metal-rich ([Fe/H] > -1.0) regime, but underperform in the metal-poor regime compared to MLPs trained directly on LRS. We also show that the optimal fine-tuning strategy depends on the specific stellar parameter under consideration. These results highlight that simple pre-trained MLPs can provide competitive cross-survey generalization, while the role of spectral foundation models for cross-survey stellar parameter estimation requires further exploration.
format Preprint
id arxiv_https___arxiv_org_abs_2602_15021
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI
Zhao, Xiaosheng
Ting, Yuan-Sen
Wyse, Rosemary F. G.
Szalay, Alexander S.
Huang, Yang
Dobos, László
Budavári, Tamás
Wei, Viska
Solar and Stellar Astrophysics
Astrophysics of Galaxies
Machine Learning
Cross-survey generalization is a critical challenge in stellar spectral analysis, particularly in cases such as transferring from low- to moderate-resolution surveys. We investigate this problem using pre-trained models, focusing on simple neural networks such as multilayer perceptrons (MLPs), with a case study transferring from LAMOST low-resolution spectra (LRS) to DESI medium-resolution spectra (MRS). Specifically, we pre-train MLPs on either LRS or their embeddings and fine-tune them for application to DESI stellar spectra. We compare MLPs trained directly on spectra with those trained on embeddings derived from transformer-based models (self-supervised foundation models pre-trained for multiple downstream tasks). We also evaluate different fine-tuning strategies, including residual-head adapters, LoRA, and full fine-tuning. We find that MLPs pre-trained on LAMOST LRS achieve strong performance, even without fine-tuning, and that modest fine-tuning with DESI spectra further improves the results. For iron abundance, embeddings from a transformer-based model yield advantages in the metal-rich ([Fe/H] > -1.0) regime, but underperform in the metal-poor regime compared to MLPs trained directly on LRS. We also show that the optimal fine-tuning strategy depends on the specific stellar parameter under consideration. These results highlight that simple pre-trained MLPs can provide competitive cross-survey generalization, while the role of spectral foundation models for cross-survey stellar parameter estimation requires further exploration.
title Generalization from Low- to Moderate-Resolution Spectra with Neural Networks for Stellar Parameter Estimation: A Case Study with DESI
topic Solar and Stellar Astrophysics
Astrophysics of Galaxies
Machine Learning
url https://arxiv.org/abs/2602.15021