FWIW I'm sympathetic to vibe-coded docs as I'm doing it myself a bit lately, but the agents are bad at it by default because all their context is the how and why of technical decisions made while coding with you
replythey need specific coaching to get them to try to write for the perspective of a new user
The main reason to use it is the output quality. SRT steers the model toward a consistent target voice or discourse style more reliably than prompting or basic steering, while keeping the base model frozen. The results feel more coherent in tone and perspective across longer outputs, especially when the target style comes from a specific corpus or community.
On the sympathetic point about vibe-coded docs: exactly.
replyhow is it different/better than LoRA ?
replyThanks for the feedback … rough and precise equally appreciated. Computational semiotics was empirically proven with this repo. I will work hard to make the findings and content more accessible for everyone.
reply