This is definitely a neat problem.
Just thinking about it, I'd break the contribution of the weights into two categories.
1) changing the centers of gravity and percussion. The best way to think of this is as a sliding weight that can be fixed in place. Fix it at the bottom of the slide, and the sword acts more as a back-weighted stabbing weapon. Fix the weight forward, and the sword acts better as a forward-weighted chopping weapon.
2) the sliding weight acts to increase the impact force by some amount. This gets weird. For instance, the weights could add a second impact to a stab. You use the blade to stab forward, and once the blade slams to a halt, the weights slide forward and add a second thump.
Conversely, when sliding, the weights move in an arc (potentially a straight line) out, from resting near the hilt to the outer end of the slide as the blade moves. Again, they add some force at the end of the slide, but it is at an angle to the cut. This might or might not be a good thing.
In both cases, there's a notable jerk or twitch when the sliding weight hits hits either end of the slide. This might or might not be a good thing.
The nice thing is that this would be relatively easy to dummy up. Put ball bearings (or whatever) in a closed pipe of the length of the slide, and attach this pipe to a practice blade (perhaps with duct tape?
). Some empirical experimentation would be informative, I think.