That's mostly a really nice explanation, the one thing that bugged me (which is my pet peeve in math-y CS papers) is a not-perfectly-clear explanation of your v

That's mostly a really nice explanation, the one thing that bugged me (which is my pet peeve in math-y CS papers) is a not-perfectly-clear explanation of your variables. The ones in the equations interspersed with the diagrams of the original derivation are easy enough to pick up from the diagrams, but you switched from $n$ to $l$ for the number of inputs in the correct expression (and the Bose paper also used $n$, so I'm not sure why).