Over the past decade, I have been deeply involved in the development of a comprehensive approach for the theoretical description of neutrino-nucleus interactions---which will be indispensable to reduce the systematic uncertainty of accelerator-based searches of neutrino oscillations---using the formalism successfully employed to study electron-nucleus scattering. The first important result of this effort has been the generalisation of the factorization scheme to the description of neutrino-nucleus cross sections in the kinematical region in which the impulse approximation is expected to be applicable [Phys. Rev. D 72 (2005) 053005; Nucl. Phys. A 789 (2007) 379]. These studies contributed to expose the inadequacy of the relativistic Fermi gas model of nuclear dynamics, which is still routinely used in simulation codes.

Based on the analysis of the double differential neutrino-carbon cross section in the charged-current quasi elastic sector, measured by the MiniBooNE Collaboration, in 2010 I have first argued that the failure of most existing theoretical models to simultaneously explain electron- and neutrino-nucleus data must be ascribed to the uncertainties associated with the flux average procedure, hindering the identification of the dominant reaction mechanism in neutrino interactions [PRL 105 (2010) 132301]. I have discussed this issue, as well as the need of a new paradigm, suitable for the description of processes in which the lepton kinematics is not fully determined, in a plenary talk delivered at the XXIV International Conference on Neutrino Physics and Astrophysics (Neutrino 2010, Athens, June 14-19, 2010). Work in this direction is currently being done. Early results include the extension of the factorization scheme to the treatment of inelastic channels [PRL 118 (2017) 142502] and processes involving two-nucleon meson-exchange currents [PRC 99 (2019) 025502].

In order to oversee the implementation of the spectral function formalism in the universal neutrino event generator GENIE, between August 2013 and March 2014 I have spent a semester of sabbatical leave at the Centre of Neutrino Physics at Virginia Tech (Blacksburg, Virginia, USA). The results of this work [Phys. Rev. D 90 (2014) 093004] have been extensively and successfully validated through comparison to electron scattering data.

In 2014, I have also proposed a measurement of the electron-argon cross section at beam energy of 2 GeV, to be performed at the Thomas Jefferson National Accelerator Facility (Jefferson Lab, Newport News, Virginia, USA). This experiment is meant to provide a most valuable input for the ongoing and future studies of neutrino interactions in liquid argon detectors. The proposal, that I have successfully defended before the Jefferson Lab Program Advisory Committee in July 2014, has been approved for the full amount of requested beam time, with an excellent scientific grade (A-). Data taking has been completed in March 2017, and the analysis is under way [Phys. Rev. C 98 (2018) 014617, 99 (2019) 054608, 100 (2019) 054606,

As an important by-product of the application of the factorization scheme ad the spectral function formalism, I have been able to obtain accurate estimates of the emission rate of γ-rays produced in the aftermath of neutral current neutrino- and antineutrino-oxygen scattering processes [PRL 108 (2012) 052505]. The availability of this information is required for the analyses of neutral current interactions in water Cherenkov devices, which are used to detect supernova neutrinos with average energy between 20 and 30 MeV. The results of my calculations turned out to be compatible with the measurements recently reported by the K2K Collaboration.

I have recently co-authored a review paper on the modelling of neutrino interactions, and its impact on the determination of the oscillation parameters [Phys. Rep. 700 (2017) 1]