With somebody that is about as advanced as you, you can bounce off ideas, have interesting discussions, and learn a lot. And then got discuss with the professor.
Another problem with working with world class researchers (as a grad student) is that you will not learn the most important part of research. The hardest part of research is finding questions, not answers. Good questions that is. Questions that are interesting and unsolved, but solvable. This is very hard. If you work with someone that is really very strong he will have more ideas than time to investigate them. So you will at best become good at solving problems, but you will not learn how to find problems. In mechanical_fish's comments, he says that most researchers "publishing over-complicated solutions to easy problems, or non-solutions to difficult problems, or incomprehensible solutions to niche problems". This not because they are dumb or lazy, it's because they can't do the hardest thing in research: find good questions.
Another thing (that others have already pointed out): expect failure. Good research is hard. In fact even bad research is hard. Doing new things is hard, doing interesting things is hard, and doing new interesting thing is incredibly hard. So most of what you try will end in failure. Get used to it or get out of academia as quickly as possible. Not many people have the right mental constitution to live a happy life while they continuously fail. Look around you; you will probably see many people that are not happy with life. I think this has something to do with it.