Does the LLM option not help to actually parse the text?

I'm getting tons of mis-parsing of paragraphs with Marker. Some paragraphs are broken into two paragraphs for no reason, and some paragraphs are combined for no reason despite each being indented.

I thought the LLM option could help with this, since without that Marker doesn't seem to do much different from traditional OCR software from 15 years ago. But it doesn't seem to make any difference at all