Even if you are a graduate student, you may have to be very independent because your adviser is too busy or not knowledgeable about what you want to study. It is not necessarily a bad thing. I had a very caring Ph.D. thesis adviser, but I wondered from time to time if he took too good care of his students.
What a good adviser can do for you is to pick a good research topic. Even if your professor handed you a research topic, you should still go to “Web of Science” for literature research. WoS is not perfect, because it will not tell you what people are writing since it takes a year or two for a paper to come out. If your adviser goes to many conferences, he or she should know who you are competing with. Competition is not always a bad thing, if you can beat the other guys at the finishing line.
If you are doing ocean modeling like I do, your adviser should also help you to pick a “good” model. None of the numerical models are perfect, but some are better than others. You need the right tool to solve a particular problem, like using the correct key to open a particular door.
Once you have started your project, you can reach out to experts in the field. If you have a question about a particular paper, email the lead author. (Here comes the language issue. Most scientific exchanges are carried out in English, at least in my field, so sharpen up your English skill. If you have a question about English, drop me a line at my Blog site. I may not have the answer for you right away, but I will try to find the answer for you.)