Spatio-Spectroscopic Representation Learning using Unsupervised Convolutional Long-Short Term Memory Networks

Integral field unit (IFU) spectroscopy provides spatially resolved spectra across galaxies, offering crucial insights into their evolution. However, its high observational cost limits current IFU datasets to $\sim 10^4$ objects. We present a multi-modal, probabilistic foundation model that predicts high-resolution spectra with calibrated uncertainties at arbitrary spatial locations within a galaxy directly from broadband images. Built on a masked autoencoder framework, our architecture injects fiber positional encodings and redshift aware wavelength encodings, enabling spatially conditioned predictions. Trained on 4.7 million images and single fiber spectroscopic observations from the Dark Energy Spectroscopic Instrument (DESI) survey, our model exploits the natural variance of fiber placements and the morphological self-similarity of galaxies to achieve IFU-like capabilities without any IFU training data. Predicted emission line flux maps match independent IFU observations from the Mapping Nearby Galaxies at APO (MaNGA) survey, with performance comparable to a supervised baseline trained directly on IFU data.

Automated characterization of galactic substructure is an essential step in understanding the transformative physical processes driving galaxy evolution. In this study, we investigate the application of deep learning (DL) frameworks to characterize different galactic substructures hosted within parametric light-profile subtracted residual'' images of a large sample galaxies from the CANDELS survey. We develop a supervised Convolutional Neural Network (CNN) and unsupervised Convolutional Variational Autoencoder (CvAE) and train it on the single-Sérsic profile fitting based residual images of $10,046$ bright and massive galaxies ($H<24.5\,{\rm mag}$ and $M_{\rm stellar} \geq 10^{9.5}\,M_{\odot}$) spanning $1<z<3$, in conjunction with their visual-based classification labels indicating the nature of residual substructures hosted within them. Using our unique data preprocessing approach, we prepare our residual images such that the inputs to our DL networks comprise only galaxy of interest'', and augment them such that our sample span uniformly across different residual characteristics. We assess the latent space of the CNN and CvAE using Principle Component Analysis (PCA) along with independently quantified metrics of residual strength (significant pixel flux $SPF$, Bumpiness, and Residual Flux Fraction). We also employ an unsupervised Gaussian Mixture Modeling (GMM) based clustering scheme with Support Vector Classification (SVC) to identify groupings in PCA space that correspond to similar residual substructure. We find that our supervised CNN latent features in PCA space correlate with the $SPF$ values and distinguish between qualitatively strong and weak residual substructures. While our unsupervised CvAE latent space also correlates with visual and quantitative residual characteristics, but lacks clear discriminatory power when characterizing different residual substructures.

Auroral spectrographs such as the Auroral Spectrograph In Skibotn (ASIS) record hundreds of thousands of emission spectra, but only a few hundred can be labelled by an expert. To exploit the rest, we pretrain a 1D Vision Transformer with a masked autoencoder on 223,000 unlabelled spectra. Without labels, its representation recovers the emission-line intensity ratios that physicists use to diagnose the precipitating particles (R^2 0.91 vs. 0.77 for an untrained control) and, under one linear probe, classifies as well as 13 features designed by experts. Fine-tuned, the model outperforms the previous supervised auroral classifier on its own benchmark (macro-AP 88.5 vs. 77.8), reaches 0.870 mAP, and exceeds the same architecture trained from scratch by +0.159 with 10% of the labels; attribution shows that it uses both N2+ bands. Could an existing pretrained model replace it? Two astronomical spectral foundation models and a time-series model transfer according to their spectral window: SpectraFM, trained in the infrared, falls below the untrained control, whereas SpecFormer, trained in the optical, approaches in-domain pretraining without reaching it.