EvtSLBaryonAmp class used by some Lb2LpnuX models sets internally spin-density matrix, which makes little sense and also leads to problems in case we try to decay particle in rest in LAB frame. We will need to review what everything is using this and potentially clean it up in all relevant models.
Possibly related to email thread "Potential weird EvtGen bug in semileptonic baryon decays(?)" from @florian et al.?