Rubino et al call their mechanism "amplification by scattering from a relativistic inhomogeniety" an, although I am not a quantum optics guy at all, I will explain the mechanism that I beleive is in place. These two papers underlie what are now called "golf rumors:"
Negative frequency resonant radiation
https://arxiv.org/abs/1201.2689v1Soliton-induced relativistic-scattering and amplification
https://arxiv.org/abs/1211.0256Lasers are very simple compared to light bulbs because the photons are monochromatic and in phase. By crossing laser beams, you can create surfaces of constant phase. The "phase" in a laser beam is (basically) the electric field (becuase light is EM radiation) and you can set up a system of crossed laser beams so that you get a surface in the intersection where there is an E=0 surface of phase lock despite the phase of the individual beams oscillating. In classical EM, however, E=0 is the boundary condition which describes the surface of a piece of metal. This surface acts like a virtual metal foil in the laser beam, and I think they can do a virtual (inverse) photoelectric effect with it to get free energy photons. You can see that that E=0 surface is "solitonic" because it it is the absence of the E field, and it is a relativistic inhomogeneity because the null field surface is bascially surfing in the intersection of laser beams. I describe the effect as "stimulated emission from the vacuum" which is like the second quantized version of stimulated atomic emission. Since the virtual foil in the crossed lasers is 2D, it has two equal but oppositely signed normal vectors. I think the photons shoot out on the back side of the 2D virtual where they pick up an extra minus sign from the backwards normal vector so that the negative frequecy modes have positive, physical energy in the lab frame.
https://www.youtube.com/watch?v=wyKQe_i9yyo&t=47