Removing Dust from CMB Observations with Diffusion Models
This work addresses the challenge of refining dust foreground models for cosmology, specifically to aid in detecting primordial B-modes in CMB data, representing an incremental improvement in component separation techniques.
The paper tackled the problem of separating Galactic dust foreground from cosmic microwave background (CMB) observations using diffusion models, showing that these models can recover dust and CMB components with accurate summary statistics like power spectra and Minkowski functionals, and introduced a cosmology-conditioned model that outperforms single-cosmology training.
In cosmology, the quest for primordial $B$-modes in cosmic microwave background (CMB) observations has highlighted the critical need for a refined model of the Galactic dust foreground. We investigate diffusion-based modeling of the dust foreground and its interest for component separation. Under the assumption of a Gaussian CMB with known cosmology (or covariance matrix), we show that diffusion models can be trained on examples of dust emission maps such that their sampling process directly coincides with posterior sampling in the context of component separation. We illustrate this on simulated mixtures of dust emission and CMB. We show that common summary statistics (power spectrum, Minkowski functionals) of the components are well recovered by this process. We also introduce a model conditioned by the CMB cosmology that outperforms models trained using a single cosmology on component separation. Such a model will be used in future work for diffusion-based cosmological inference.