LLM reinforcement learning improves those tasks by having the agents "reason" until they hit a verifiable output state that passes a black or white test. So they learn language which is tuned to the task in a silo, never communicating with anyone or anything.
In contrast, most human communication and writing is intended for other humans, and we constantly hit failures here that we learn to iterate on and improve. Humans who do math do so with the constraints that their math needs to be reviewed by another human and in the context that all their writing needs to be understood by another human.
The process causes a warped and compressed version of language.
I updated my vibe config to point to Mistral 4 and it charges my subscription capacity, not my API allowance. So it is included in subscriptions, you just need to do a little bit of legwork.
How is the primary way of interacting with your computer being good a gimmick?
I try to use my partner’s windows laptop and it’s borderline unusable for me. She only almost uses it with a mouse plugged in. Meanwhile I don’t think I’ve ever seen someone plug in a mouse to a MacBook that wasn’t docked to a monitor and keyboard.
Similar experience, the first thing it did for me was update production data models without a confirmation. It has always shown me the command is going to run and then I confirm.
In contrast, most human communication and writing is intended for other humans, and we constantly hit failures here that we learn to iterate on and improve. Humans who do math do so with the constraints that their math needs to be reviewed by another human and in the context that all their writing needs to be understood by another human.
The process causes a warped and compressed version of language.
reply