> It would not be a problem if Arxiv was full of such articles, that would never cut it in an actual publication, but such articles are mixed with many that are eventually published and that just adds an amount of noise to the process that makes it hard to be sure what you're reading
I don't see this is a problem as, if you want to cite an arXiv paper, the responsibility on you increase, to ensure that the methodology etc is sound. This would be a better state of affairs than the "you must cite this highly cited paper just because it's a highly cited paper" situation we're in now.
Or by someone associated with them (in their lab, uni, co-authors as you mention etc). Essentially this means you must cite certain papers in some areas, as for larger conferences you're sure to be reviewed by such people.
I don't see this is a problem as, if you want to cite an arXiv paper, the responsibility on you increase, to ensure that the methodology etc is sound. This would be a better state of affairs than the "you must cite this highly cited paper just because it's a highly cited paper" situation we're in now.