My standard for “good” practise is: if everyone adapt said practice, then the world would be better off.
Even though the effort of a single person can be futile; if I cannot chance my behavior for a cause I believe in, how can I expect the rest of the world to do the same?
My conspricy theory is that early LLMs have a hard time figuring out the logical relation between sentenses, hence do not generate good transitions between sentences.
I think bullet point might be manually tuned up by the developers, but not inheritly present in the model; because we don’t tend to see bullet points that much in normal human communications.
Hum, you will need to transition first…