A quick read of the paper shows it's mostly focused on systems with discrete levels; IIRC there's already plenty of thermodynamic weirdness in those, like "nega

A quick read of the paper shows it's mostly focused on systems with discrete levels; IIRC there's already plenty of thermodynamic weirdness in those, like "negative temperature" states in the Ising model and so forth. They touch on a continuous system in the final section, but that also includes a system with a finite upper bound to its energy (phonons). Wonder if that's related to the potential for 2nd Law violations.