Repeatedly applying the question "can we leverage <computer hardware subsystem> to exfiltrate data?" is valuable research.
The author is clearly an expert in exfiltration and has deep insights in the area. But instead of working on presenting the big picture, prefers to throw breadcrumbs to the community. After a dozen of "exfiltration using X" papers, one would at least expect an SoK paper.
If the answer was "we tried really hard, but couldn't figure it out" pretty sure the paper would not get published.