# CosmoFlow: Scale-Aware Representation Learning for Cosmology with Flow Matching

arXiv: https://arxiv.org/abs/2507.11842 · HTML: https://mlsanity.com/paper/2507.11842/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Inverse problems and image restoration, one of 14 directions in Diffusion models.
- Builds on 13 of its 20 references that are in the map.
- First posted on arXiv 16 Jul 2025.

## Placed in

- Diffusion models / [Inverse problems and image restoration](https://mlsanity.com/diffusion/inverse-problems-and-image-restoration/)

## Builds on (most cited first, 13 of 13)

- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Score-Based Generative Modeling through Stochastic Differential Equations](https://mlsanity.com/paper/2011.13456/) (2020)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Decoupled Weight Decay Regularization](https://mlsanity.com/paper/1711.05101/) (2017)
- [Stochastic Interpolants: A Unifying Framework for Flows and Diffusions](https://mlsanity.com/paper/2303.08797/) (2023)
- [Variational image compression with a scale hyperprior](https://mlsanity.com/paper/1802.01436/) (2018)
- [End-to-end Optimized Image Compression](https://mlsanity.com/paper/1611.01704/) (2016)
- [Flow Matching Guide and Code](https://mlsanity.com/paper/2412.06264/) (2024)
- [SODA: Bottleneck Diffusion Models for Representation Learning](https://mlsanity.com/paper/2311.17901/) (2023)
- [Exploring Diffusion Time-steps for Unsupervised Representation Learning](https://mlsanity.com/paper/2401.11430/) (2024)
- [Multifield Cosmology with Artificial Intelligence](https://mlsanity.com/paper/2109.09747/) (2021)
- [The CAMELS project: Expanding the galaxy formation model space with new ASTRID and 28-parameter TNG and SIMBA suites](https://mlsanity.com/paper/2304.02096/) (2023)

## Abstract (arXiv)

Generative machine learning models have been demonstrated to be able to learn low dimensional representations of data that preserve information required for downstream tasks. In this work, we demonstrate that flow matching based generative models can learn compact, semantically rich latent representations of field level cold dark matter (CDM) simulation data without supervision. Our model, CosmoFlow, learns representations 32x smaller than the raw field data, usable for field level reconstruction, synthetic data generation, and parameter inference. Our model also learns interpretable representations, in which different latent channels correspond to features at different cosmological scales.