Oddly enough, it does work quite well when inverted.

But it depends a LOT on the airplane, and the airfoil, which is why the first thing I did was ask what the airplane is. If it has an airfoil with 0-camber or nearly-0, then reflex may not help, or it may hurt as you say. But if it's a design with a bunch of camber, some reflex can do wonders.

I still think the problem sounds most like a CG problem, but I also think there is not enough information provided to really know what is going on.

Other info that would be nice to know:

- Is it a gradual pitch change or very sudden.
- The post mentions slow speed, does that mean it only happens on final approach to landing?
- After the pitch change, what happens? Does the nose flop down, or does a wing drop, or does the plane just continue to climb?
- What happens if you are flying full throttle and trimmed level, then suddenly pull off the power?
- What happens if you climb way up, get level, then cut the power to idle or even kill the engine and then enter a 45degree dive?

The results of this kind of testing will tell you if the problem is related to airspeed, engine thrust, or something else.

For example, if the plane is flown around at full throttle all the time, and the only time the pitch occurs is when the power goes from full-on to full-off on final, and the change is sudden, but isn't followed by a stall, it might very well be the thrust line as Pe Reivers suggested.