pull down to refresh

Almost every time I read an abstract for a paper on arxiv (multiple times per day), I'm having an underlying feeling that it's "sloppy", either of: written by an LLM, rewritten by a human from an LLM's template, or translated through LLM.
I often need to edit the abstract for readability when I decide to post it here.
So the 30% AI+Unclassifiable is lower than I thought it would be.
Big problem here: I'm not a native English speaker, so I find it hard to detect sloppy AI.
reply