Sounds like you need to tweak the PID gains on the boost control output. Does Link use a feedforward strategy where you specify a base boost control output duty cycle, and then a PID algorithm trims it on top of that? That's the most common method I've seen, and if that's the case, you might need to tweak the base boost duty cycle by turning off PID feedback for boost control and doing some pulls and seeing what the base duty cycle gives you. With that dialed in, tweaking the PID variables will give you much more control over different behavior of the boost output. The D gain specifically tries to calm the system moving quickly past the setpoint. You might have too much P gain and not enough D.
But I'll be honest, some standalones just don't have the best control algorithm implementation. I could not get my MS PNP2 to give good closed loop control in all circumstances, with a system that seemed both slow to respond and very sensitive to adjustments like it had a long refresh time but used high background gains to try to overcome that. I gave up after experiencing low RPM boost overshoot like you are seeing on your setup. My AEM Infinity has excellent boost control response though, with a super fast refresh time on the PID algorithm.
Also, I'd raise the boost cut a bit. No reason to run it just a few psi from your target boost, as you get into situations where small amounts of overshoot trigger it.