The whole point of an FRO is to have less oversight than a traditional government grant or contract.

That is the point, but not the purpose. Traditional funding mechanisms leave large gaps across whole research areas. A more traditional government grant is small and focused, and only makes sense within the context of an ongoing research effort, e.g. in a university. That's great and fine, except that there are whole categories of things that existing labs and agencies do not do, yet are still too close to basic science for an industrial research lab to pick up.

For example, let's say you want to make a fundamentally different kind of microscope, or nano-fabrication lab. Something that requires many millions of $$ to setup, and further tens of millions to operate. That's too much for an SBIR grant, too expensive for industry to do on a lark. An FRO could setup the lab and run it for 5-7 years, accomplishing major research goals while proving out the concept. But it'll never get done on a spoon-fed sequence of small grants.

So you grant an organization $50M/yr to do this work, with broad freedoms to spend that money how they see fit, so long as progress is made towards the research goals. That's an FRO. Yes, technically that is a government grant with less regulatory oversight. But it's not libertarian philosophy that drives that outcome, so much as regulatory oversight is not setup to fund such efforts in the first place.