Ah, the paper itself is much better than I'd expected
It's also unwieldy - looks more like a PhD thesis.
(bloody shame we can't just have an internet that delivers the information concerned rather than having to wait for someone to sleuth down the actual dirt)
They do indeed define "populism" in section 2.1, rather thoroughly, as political "splitting" (I borrow a psych term here).
It's actually filled with rather good analogies.
A keyword they use a lot is "self-destruct". And yes, giving people "what they want" (rather than "what we all need together") is unsustainable short-termism.
Haven't time to read the full 190 pages, but my scepticism about the solidity of the key concepts remains. For example in the methodology:
"For each of the 1,458 leaders in our sample, we assign the value of 1 if the leader is a populist, and 0 if the leader is not a populist (non-populist)."
It's a critique of splitting that is itself predicated on a binary partition.