P3724R3 proposes Constant-checked preconditions to make requirements for some library UB clearer and more consistent. It seems to me that switching to say Constant-checked preconditions can be editorial, and it might be useful for signaling future library papers to consider which kinds UB should be diagnosed during constant evaluation.
Can we editorially add that element before considering the rest of that paper? CC @eisenwave @jwakely.