Some time ago, the following issue existed in triple quads:
The triple would start measuring transition 1: mass A --> mass B
The first quadrupole would be set to mass A, the final quadrupole to mass B, and you'd measure for some msec.
Then the instrument would change to transition 2: mass C --> mass D
Now, at the instant of change, there may still be ions flowing through the instrument, so it's possible some left-overs of mass A will be passing through the collision cell, and generating fragments, some of which might be able to pass through the final quadrupole even though it's now looking at mass D (especially if you're working on a number of related compounds that will have similar fragments), and there may even be some ions of mass B that are far enough through the final quadrupole, or between it and the detector, that will still show up.
The ability of ions from one transition to be detected in the early part of the next transition's dwell time was called cross-talk. It was minimised by setting an appropriate delay between the two transitions, long enough to clear all ions from the system.
I believe this sort of cross-talk has been completely eliminated in modern instruments.