In my opinion, an increase in 1% in the score is not a “substantial” change in performance.
I, and perhaps my colleagues, agree that this last question could be rephrased in a way that would make it more understandable
Notice that the aim of this exercise is to compare performance across folds and not the mean accuracy.
But I agree that maybe we can rephrase for the next session using the word “significantly” instead of “substantial”. The former as a synonym for “meaningful” captures better the idea of a systematic improvement (deterioration) across folds.