The PhD Metagame: Don't Make Things Actually Work maxwellforbes.com 6 points by teleforce 12 hours ago
turtleyacht 12 hours ago > all of your work would now be irrelevantWouldn't prior papers be cited as known avenues to rule out? The older papers would turn up in searches.With AI, are we no longer reverse-engineering the learning model to retrieve specific heuristics?It may conserve power to convert the simulated solution from the prompt to a set of analytic calculations, maybe.
> all of your work would now be irrelevant
Wouldn't prior papers be cited as known avenues to rule out? The older papers would turn up in searches.
With AI, are we no longer reverse-engineering the learning model to retrieve specific heuristics?
It may conserve power to convert the simulated solution from the prompt to a set of analytic calculations, maybe.