You mentioning "homework" makes me think that you are possibly in BSc or MSc programme, but not PhD yet (although, you sound like you're heading there). So, I'll try to concentrate my answer on advice useful to somebody just starting off as a (potential) researcher. Sorry on long text, but since the answer is based on experience it wouldn't have any value without an explanation.
I, personally, would have loved if my supervisor from my first standalone bigger project was micromanaging me. Moreover, I explicitly told the guy "I'm very new at this, and would appreciate instructions and directions on how to approach such a big project efficiently". He left me more-or-less alone, I was a beginner, and didn't handle the situation very well: but one important lesson was that that specific type of adviser was not a good match for me.
On the other hand, the professor that was my primary adviser during my BSc and MSc was a very smart, eloquent and confused scientist. When I first told him I am considering continuing with academia after MSc, very often I would get a feeling he was paying way too much attention to "useless" details, such as:
- asking me to understand the literature to impossible details
- making me justify using (sub)methods, (sub)approaches and other sub-parts of approaches he made me study, which were chosen by the authors of the papers he suggested for reading
- correcting and re-writing my written works as one non-native English speaker to another, where I was sure my grammar was fine
- making a loud point of small spelling mistakes in Drafts V.001
What helped was, as most people here suggested: talking to him. He got to explain to me that I have to be able to argument my choices, not by saying (cf. John Doe: "Important work on the topic"), but by properly explaining the idea behind the approach, and then properly explaining why it is applicable to my situation. He explained that, he, as an expert, does not know answers to all questions, solutions to all problems, but is in a good position to give an educated guess. A bigger picture here is that if you learn how to express yourself clearly and with arguments, the scientific community can understand you, and build on your work. Bottom line, they can help developing your idea. In the end, in academia today, it is not enough to have a good idea, you have to explain your idea to everyone, and it's surprisingly hard to do even after you feel like you understand everything.
So, from that point of view, what he was doing was actually:
- teaching me critical thinking, and how to properly read scientific work
- teaching me that just because something is published, does not make it correct
- teaching me that just because something grammatically correct, does not mean it is clear, simple and understandable: and understandable is always your goal
- teaching me that everyone looks down on lazy-asses who don't spell-check their work. It's not nice, but that's how it is :/
In return, since he got my critical thinking working, he would do the same for me: explain and argument his choices and recommendations, tell me why something might not seem important now but will be good in my field later, hinting me on his methods to read scientific literature, explain before recommending literature what he is expecting me to learn from it, etc.
The bottom line is, from the professors side, I think it is very hard to develop a good "scientific style" in somebody, unless you're imposing your own. It's different with somebody with experience and a (semi-)established style, it's different (and possibly more difficult) when starting with somebody new to it all.
And, from student side: If you're still advancing, doing good work, and learning, it's not bad. If you're just in the beginning and will soon change supervisor anyway (e.g. from MSc to PhD), it's even better: you get expirience working with different kind of people, and it will be easier for you to find an adviser with a style compatible with yours.
In the end, it all comes down to communication: if you are on good terms with your adviser, just the explanations on why she's recommending what she is carry invaluable information that assimilate for years. If it's not just mindless task-giving, it teaches you things in very fast and simple (although sometimes annoying) manner