Higgs couplings: disentangling New Physics with off-shell measurements

After the discovery of a scalar resonance, resembling the Higgs boson, its couplings have been extensively studied via the measurement of various production and decay channels on the invariant mass peak. Recently, it has been suggested the possibility to use off-shell measurements: in particular, CMS has published results based on the high- invariant mass cross section of the process $gg to ZZ$, which contains the contribution of the Higgs. While this measurement has been interpreted as a constraint on the Higgs width after very specific assumptions are taken on the Higgs couplings, in this letter we show that a much more model-independent interpretation is possible.

Searching for a lighter Higgs: parametrisation and sample tests

The structure of the Higgs sector is a major issue in the quest of a detailed description of the electroweak interactions. Most of the effort is devoted to the study of the standard model--like Higgs boson at 126 GeV, however the experimental collaborations at the LHC are also searching for extra scalar particles whose presence may hint to an extended Higgs sector, typical of many extensions of the standard model. We study a model independent parametrisation of a scalar particle lighter than the 126 GeV Higgs boson, which may be easily implemented in the ongoing searches by ATLAS and CMS. Indeed many effective Lagrangians/parametrisations used at present for the description of the Higgs sector implicitly assume that no light particles other than the standard model ones are present in the spectrum. We therefore introduce a parametrisation of a two scalars model, one corresponding to the 126 GeV Higgs boson and the other to a lighter scalar. After the introduction of such a tool, we consider two typical examples falling in this category, in order to illustrate the use of our formalism: the two Higgs doublet model and the next to minimal supersymmetric standard model. Our results agree with the specific studies performed for these models. Furthermore, employing such a generic parametrisation allows us to translate the bounds to any model beyond the standard model falling into this class.