I am unsure if there would be any benefit in Karma XPU but I could potentially think of benefits when doing GPU simulations that still require some CPU workload (or transferring collider meshes to the GPU?).
In case you don't know what it is, this is what NVIDIA writes in their blogpost [www.nvidia.com]:
What Is Resizable BAR?
Resizable BAR is an optional PCI Express interface technology. As you move through a world in a game, GPU memory (VRAM) constantly transfers textures, shaders and geometry via many small CPU to GPU transfers.
With the ever-growing size of modern game assets, this results in a lot of transfers. Using Resizable BAR, assets can instead be requested as-needed and sent in full, so the CPU can efficiently access the entire frame buffer. And if multiple requests are made, transfers can occur concurrently, rather than queuing.
If there is any benefit I would probably upgrade my whole System, if there isn't then I doubt upgrading my CPU right now would give me any noticeable performance increase as it has got more than enough cores and speed.
Maybe somebody who does have a System that supports it and is interested could do a small benchmark with Karma XPU, Vellum, or the new MPM solver by running it once with it enabled and once disabled. Or maybe a Dev could also answer this question.