Off and on topic I guess but: Language and Thermodynamics? Like, the same book? That sounds interesting.

Holding symbols in a useful order durably and accessibly is an ongoing energetic event. How/why does it happen?

My old thermodynamics professor used to say: the answer's always entropy.

entropy/information theory may be the bridge?

correct, while it might not be regex it can be bypassed with regex. They do have a sematic classifier, but it's really weak on opus 4.8 and (was) weak on fable, but they either added a lot more regex strings or the classifier is actually good now.

Try doing what congress does: take a bill from the house, gut it and put in what you want after the house passes it