Alternatives to $\Lambda$: Torsion, Generalized Couplings, and Scale Invariance

We present a comparative analysis of current observational constraints on three recently discussed alternative models for explaining the low-redshift acceleration of the universe: the so-called steady-state torsion model, the generalized coupling model, and the scale invariant model by Maeder (an example of a broader class which we also briefly study) These are compared to the traditional parameterization of Chevallier, Polarski and Linder. Each of the candidate models is studied under two different assumptions: as genuine alternatives to $\Lambda$CDM (where a new degree of freedom would be expected to explain the recent acceleration of the universe without any cosmological constant) and as parametric extensions of $\Lambda$CDM (where both a cosmological constant and the new mechanism can coexist, and the relative contributions of both are determined by the data). Our comparative analysis suggests that, from a phenomenological point of view, all such models neatly divide into two classes, with different observational consequences.