Celia: Catalog Embeddings Learned from cosmologIcal simulAtions
Abstract
Cosmological hydrodynamical simulations connect observed galaxies to cosmology and the physics of galaxy formation. Each simulation is expensive, and different codes model galaxy formation in different ways. We present Celia (Catalog Embeddings Learned from cosmologIcal simulAtions), a self-supervised foundation model that maps catalogs of subhalos and their galaxy properties to an embedding vector for the purposes of compression and organization of astrophysical and large-scale structure information. Celia is pretrained without parameter labels on the IllustrisTNG, SIMBA and Astrid suites of CAMELS and their N-body counterparts, by reconstructing masked subhalo properties and predicting the representations of a teacher network. Simple linear models fitted to the embeddings recover Ωm with R2=0.999, and σ8 and both supernova-feedback amplitudes with R2≥0.97, in all three suites. The embeddings translate well to out-of-distribution simulations such as Swift-EAGLE, a code that Celia never saw. A linear model fitted on the three training codes predicts its Ωm with a correlation coefficient r=0.985, and linear models fitted on 20 labeled simulations reach r≥0.94 for Ωm, σ8 and a supernova-feedback amplitude. We decompose the variation among the simulations of each suite into principal components of the embedding, and simple power laws of the simulation parameters describe the leading components. Celia gives simulated galaxy catalogs a shared coordinate system for simulation-based inference, emulation, the comparison of galaxy-formation models, the detection of physical degeneracies and the discovery of new astrophysical probes.
Related
- Rates of tidal disruption events from constrained cosmological simulations of the local Universe: population properties and implications for transient surveys
- Coupling between stellar and HI lopsidedness in Milky Way--type galaxies from the Auriga Superstars cosmological simulations
- Bridging perturbation theory and simulations: initial conditions and fast integrators for cosmological simulations
- Agentic Systems for Calibrating Cosmological Simulations
- OpenCosmo: Community Portal and Analysis Framework for Flagship Cosmological Simulations
BibTeX
@misc{zhang2026celia,
title = {{Celia: Catalog Embeddings Learned from cosmologIcal simulAtions}},
author = {Xiaowen Zhang and Matthew Ho and Anshul Kumar and Tiziana Di Matteo and Rupert A. C. Croft and Barnabás Póczos},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.40218},
eprint = {2610.40218},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.40218},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Celia: Catalog Embeddings Learned from cosmologIcal simulAtions
AU - Xiaowen Zhang
AU - Matthew Ho
AU - Anshul Kumar
AU - Tiziana Di Matteo
AU - Rupert A. C. Croft
AU - Barnabás Póczos
PY - 2026
DA - 2026/10/29
PB - arXiv
DO - 10.48550/arxiv.2610.40218
UR - https://arxiv.org/abs/2610.40218
AB - Cosmological hydrodynamical simulations connect observed galaxies to cosmology and the physics of galaxy formation. Each simulation is expensive, and different codes model galaxy formation in different ways. We present Celia (Catalog Embeddings Learned from cosmologIcal simulAtions), a self-supervised foundation model that maps catalogs of subhalos and their galaxy properties to an embedding vector for the purposes of compression and organization of astrophysical and large-scale structure information. Celia is pretrained without parameter labels on the IllustrisTNG, SIMBA and Astrid suites of CAMELS and their N-body counterparts, by reconstructing masked subhalo properties and predicting the representations of a teacher network. Simple linear models fitted to the embeddings recover Ω_m with R^2=0.999, and σ_8 and both supernova-feedback amplitudes with R^2≥0.97, in all three suites. The embeddings translate well to out-of-distribution simulations such as Swift-EAGLE, a code that Celia never saw. A linear model fitted on the three training codes predicts its Ω_m with a correlation coefficient r=0.985, and linear models fitted on 20 labeled simulations reach r≥0.94 for Ω_m, σ_8 and a supernova-feedback amplitude. We decompose the variation among the simulations of each suite into principal components of the embedding, and simple power laws of the simulation parameters describe the leading components. Celia gives simulated galaxy catalogs a shared coordinate system for simulation-based inference, emulation, the comparison of galaxy-formation models, the detection of physical degeneracies and the discovery of new astrophysical probes.
ER -