I kind of struggle with my students. I always told them to put the reference number (identification) close to the cited object - to make really clear what is the cited object. For example:
- The theorem:
Theorem A  is great ...
- The study:
In the study  it is possible to see that ...
- The statement:
It was proven that usage of ... leads to problems .
However, my students keep doing something like this:
- Citation is related to the previous sentence! (probably):
... and it was proven.  The general idea ...
- As above, but with extra period:
... and it was proven. . The general idea ...
- After the entire paragraph:
... and it was proven. 
They are arguing that this referencing is correct according to the ISO norm/suggestions. They keep doing that even though I told them to not do so. And they generally try to comply with my guidelines. This is a real exception. It seems this referencing idea is deeply rooted in their brains.
- Is their way of referencing legal? If yes, where are informations how to read this fuzzy notation? And why it does not appear in top journals?
- How to tell them that this referencing is really unclear in a way that they will understand.
Note: The students I am struggling with are from all continents/countries, pursuing a university degree in electrical/software/mechanical engineering. I would say they are mostly 22±2 years old.