FAQ for Junior
Last update: 04/17/2025
I have already had several meetings with junior students (the number in left). Here are some questions that I have been asked.
- How should I schedule a meeting?
- I am a junior student. What should I do to prepare for my Ph.D. application?
- I did not get good results in my Ph.D. application. What should I do?
- I find myself not interested in research. What should I do?
- I do not know if I should go for a Ph.D. What should I do?
- I am not computer science student. What should I do if I want to do research in AI/ML?
- I found my department does not have a good research lab. What should I do?
- I find many people working on benchmark paper. Is that a good way to do research?
You can send me an email to schedule a meeting. See Calendar for available time slot. After we find a time working for both of us, you can send me a zoom invitation via schedule button. It would be great if there were a resume or webpage for me to learn about you before our conversation!
According to the current situation, I suggest you to take a summer (or not just summer) internship in person or remotely. It is a good way to find a good fit with your potential advisor. If you can have a good performance in the internship, you can get a return offer from the advisor. That is a far more stable way to get a Ph.D. position.
I am sorry to hear that. If you received any master offer that from the top universities, I suggest you to accept it and go for it. It is totally fine to take a master degree first. You can always apply to Ph.D. later with return offer, or just seek for a industry job after graduation. I do not recommend you take a gap year in China since it may get harder and harder to get the US visa, also it won't help a lot for the connection.
Not everyone has a clear research interest. Engineering job from industry is also a good choice! I recently feel that machine learning systems (MLSys) is a very promising area for engineer. You can also consider it as a good direction.
Well, it is a truth that not everyone is suitable for a Ph.D. program. Try to lead a research project just by yourself, from idea to code to paper. If you find yourself interested in research, go for it.
A fun fact is that I study civil engineering during my undergraduate. I think it is totally fine if you are not a CS student. I think it is far more important to have a strong motivation and interest. My general suggestion is keep reading papers, and try to run the code by yourself. There is no need to take courses in your college but just seek for the resources from the internet. Also, try to work with the professor in your college if you have any interest in research.
It's always a good idea to send emails to the professors (or Ph.D. students) that you are interested in in your research area. Before that, make sure you have something to show. For example, you can have some demo projects related to their current research interest.
It is important to understand why we need benchmark. Sometimes you have to build a benchmark since you are the first one defining the problem. I usually do not including only benchmark in my paper but to do with a strong and interesting baseline. However, working only on different meaningless benchmarks are not a good research.