Multi-Modal Masked Autoencoders for Learning Image-Spectrum Associations for Galaxy Evolution and Cosmology

TL;DR

This paper introduces a multi-modal masked autoencoder that learns shared representations of galaxy images and spectra, enabling reconstruction and redshift prediction with promising results for large-scale astronomical surveys.

Contribution

It adapts a transformer-based masked autoencoder to jointly embed galaxy images and spectra, demonstrating its effectiveness in reconstruction and redshift estimation tasks.

Findings

Successfully reconstructs galaxy features from heavily masked data.

Performs redshift regression with accuracy comparable or superior to existing models.

Highlights limitations in fine detail reconstruction and spectral line strength recovery.

Abstract

Upcoming surveys will produce billions of galaxy images but comparatively few spectra, motivating models that learn cross-modal representations. We build a dataset of 134,533 galaxy images (HSC-PDR2) and spectra (DESI-DR1) and adapt a Multi-Modal Masked Autoencoder (MMAE) to embed both images and spectra in a shared representation. The MMAE is a transformer-based architecture, which we train by masking 75% of the data and reconstructing missing image and spectral tokens. We use this model to test three applications: spectral and image reconstruction from heavily masked data and redshift regression from images alone. It recovers key physical features, such as galaxy shapes, atomic emission line peaks, and broad continuum slopes, though it struggles with fine image details and line strengths. For redshift regression, the MMAE performs comparably or better than prior multi-modal models in terms of prediction scatter even when missing spectra in testing. These results highlight both the potential and limitations of masked autoencoders in astrophysics and motivate extensions to additional modalities, such as text, for foundation models.