New Submission Off-forward dispersion relations for GPDs enhanced by AI and ML extrapolations

GPDs are an important tool in understanding the internal structure of nucleons; however, they cannot be accessed directly. The DVCS amplitude encodes information on the Compton Form Factors (CFFs), which are convolutions of the GPD with a hard perturbative piece. Thus, by measuring the DVCS cross section, we gain access to the CFFs, not the GPDs. Alternatively, one can measure correlation functions in lattice QCD to obtain Gravitational Form Factors (GFFs), which through Mellin moments can be inverted to give the GPDs. The GPDs have analyticity properties that satisfy dispersion relations. For off-forward kinematics, analytic continuation to unphysical kinematics is necessary. This can be accessed using AI and ML to extrapolate, as has been done with model calculations of CFFs implementing moments from Lattice calculations and limited data.



Funding Acknowledgement: DOE DNP