Comment on From Chesterton's fence to Chesterton's gapparentComments−noopprod2moI think when somebody trains code golfing LLMs with reinforcement learning they will inadvertently be smarter
Comments
I think when somebody trains code golfing LLMs with reinforcement learning they will inadvertently be smarter