If I were designing a system like this, I'd likely implement a reputation system for reporters (users who report, not journalists).

When you report content for the first time your reputation score is 0.5, if the content is of a celebrity or public figure, your reputation will invisibly drop, enough of these and your score will be so low as to not be a useful flag when evaluating whether content is objectionable.

On the other side, if the content is reviewed as violating ToS, your rep will increase.

With a system like this the fake reports and brigading will quickly lose its power, eventually all users who abuse reports will be ranked so low as to not matter.

A threshold of rep score can be set before the content is auto removed. Maybe something like 25pts in order for it to be removed, and that's after adding together all the rep of the reporting users.