"Light" here is a misnomer. What they're doing is taking the wide array of infrared wavelengths and channeling them into a single wavelength. It's easier to build a solar cell that's tuned to that wavelength.

I don't know why they're calling that "light". It's not clear to me if it means visible light. The paper seems to imply that it peaks in the near-infrared. They appear to be using "light" in the general sense of "photons"... but it's all photons.

They're talking about very high temperatures, the kind that emit significant visible light as well.