Generalization to other dark-matter annihilation channels and masses

Determine the generalized performance of the convolutional variational autoencoder signal-background separation method for dark-matter annihilation channels and dark-matter masses beyond the single 1 TeV Majorana dark-matter model annihilating through the W+W− channel used for training.

Background

The mock dark-matter analysis trains and evaluates the model using a single annihilation channel and dark-matter mass: a 1 TeV Majorana dark-matter particle annihilating into W+W−. Although the model successfully reconstructs the associated spectral feature, the paper does not establish whether this performance transfers to other annihilation channels or masses.

References

It is worth noting that the dataset only covers a single annihilation channel and DM mass, so the generalised performance to other DM channels based on this dataset remains unknown.

— Convolutional non-parametric Gamma-Ray Signal and Background Separation  (2609.30917 - Betterman et al., 25 Sep 2026) in Section 5.2, “Mock Dark Matter”