As indicated by answers to Can “whose” refer to an inanimate object? on ELU, some people wouldn't be happy with OP's use of it here. I'm not one of them, and given how awkward it would be to avoid the word in OP's construction, I'd just ignore those pedantic prescriptivist grammarians who say it's wrong.
Note specifically Peter Shor's answer on that ELU question, with several examples of Shakespeare using whose to indicate association with inanimate objects.
Here are many thousands of written instances of "equations whose solutions", which is perfectly normal English. By the same token, there's nothing wrong with OP's ...sets of elements whose pairs...
Having (I hope) dismissed criticism of whose, I'd say that most native speakers would use every rather than each in OP's particular sentence. I know I said in another answer that using every in this way is usually a bit stylised/dated. But here it's just "formal", which is appropriate in a "mathematical" context.
I can't exactly explain why "every" is better than "each" here, and I very much doubt it involves any "grammatical rule". But to support my contention that it is "better"...
whose each response (0 hits in Google Books)
whose every response (142 hits)
Also note that the corresponding figures for "whose each/every pair" are 3/116, and for ...solution they're 5/27, which may suggest that mathematicians are less attuned to such subtle distinctions.
There seems to be some disagreement over the distinction between "less common" and "ungrammatical" here, so I'd like to quote from New Scientist - 5 Aug 1989 - Page 55
Rather it elaborates a millionfold democracy whose each unit is a cell.
I've been reading New Scientist every week for decades, and I honestly can't recall ever seeing a grammatical error get through their admirable proofreading procedures. It's also reasonable to assume the authors of Encyclopedia of British Writers, 1800 to Present know their own language...
Golding is an author whose each successive work deals with a different subject in a different time...
EDIT: More recent commentsrecent comments have identified a semantic ambiguity in OP's example that I hadn't originally noticed. I assumed whose referred to "elements" (i.e. - each element within the set contains multiple words, each of which is a false friend to every other word in that element). But apparently whose references "the set", within which each single-word element is a false friend to every other element/word in the entire set.
This is purely an issue of semantics that goes beyond the issue being queried here (whether the usage whose each is "syntactically valid" in OP's example sentence).