A combination of both. This work targets multicore, but execution on the GPU is definitely possible. It's just a different line of research. I'd highly recommend taking a look at Futhark (https://futhark-lang.org/), which has a similar programming model (purely functional and parallel), but targets GPU. The compilation strategy is quite a bit different!
What I find more interesting is things like Dyna-Soar[1] as well as the parent project/paper. Futhark seems primarily very tensor-type computation specific (like APL+ extra features) and I'm interested in things that more naturally prone to code and data divergence and in trying to find ways to compensate for this.
