Abstract: Models of inelastic (or pseudo-Dirac) Dark Matter (DM) commonly assume an accidental symmetry between the left-handed and right-handed mass terms in order to suppress diagonal couplings. We point out that this symmetry is unnecessary, because for Majorana fermions the diagonal couplings are not strongly constrained. Removing the requirement of such an accidental ad hoc symmetry instead relaxes the relic density constraint due to additional annihilation modes.

In the talk, I will review the concepts around inelastic DM and introduce a simple UV-complete model realizing the new asymmetric set-up. First, I will briefly comment on how the freeze-out mechanism can produce the right abundance of DM in this class of models. I will explain how traditional constraints from (in)direct detection, beam dump experiments and colliders are modified due both elasticity and inelasticity of the model. From these constraints, we indicate two viable mass regions for the DM particle, around a few hundred MeV and around a few GeV. The former region will be fully tested by near-future analyses of the NA64 and Belle II data, while the latter turns out to be challenging to explore even with future experiments.