Academician Li Keqin

You get it, you get it; you don’t get it, you don’t get it.

How to Position Yourself

I think understanding where you belong is extremely important. In computer science, the main research directions fall into a few fairly established categories. Professor Li divides them into Theory, Algorithm, Middleware (theory-oriented), and Middleware (system-oriented). Deep learning under Algorithm is unquestionably the fashionable choice right now. An amusing situation has developed: everybody talks about AI, yet few really understand what it is. Some students choose graduate study just as blindly, entering the field before understanding it. I suggest taking a closer look at your own circumstances. What if systems research suits you better? It is hard to do something well without passion. Another group I find puzzling consists of people preparing for graduate entrance exams without knowing where they stand. They do not know whether exams suit them, or what they want to do afterward. They simply want to stay in the comfortable shelter of school and postpone employment; some even try for a second or third year. I will leave it there and wish them well.

Theory + System? No!

Wanting everything can leave you with nothing. You cannot single-handedly conduct research, file patents, start a company, capture a market, and take it public all at once. This reminds me of the many university competitions that are long on activity and short on substance. Judge their value for yourself—you probably know how much they really contribute. Building your own core capabilities is what matters.

How to Get into a Research Area

Read extensively but selectively

Know what you are doing. Do not work blindly.

Find the research topic by yourself

This is interesting. At our universities, students generally research whatever their supervisors research; choosing a topic independently is much less common. Professor Li said he selected all his own research topics while studying in the United States. Ha!

Choose a topic of your favorite

Again, it comes back to passion.

Choose a topic that matches your experience and capabilities

Again, understand where you stand.

Learn the background knowledge and skills only when necessary

This is the way it has to be.

Do not rely on your supervisor

There is a reason people say the student can surpass the teacher. Professor Li recalled that his American supervisor was an assistant professor at the time and became an associate professor soon after Li graduated. In that sense, Li helped his supervisor earn promotion. Decades later, Li kept reaching new heights, while his supervisor was still an associate professor…

Do not switch areas

Switching back and forth endlessly will get you nowhere!

How to Find an Original Idea

How do you discover something original?

Scientific research is an individual and creative process

Scientific research is a process in which an individual exercises creativity. I strongly agree with Professor Li here. He said he could not understand why papers now have five, six, or even more than ten authors. Does the original idea not come from one person? Can each person really contribute one nth of an idea? This argument might carry less weight from someone else, but Li has published more than 200 single-author papers. That is extraordinary. He joked that the other 800 or so coauthored papers among his thousand-plus publications were products of “corruption.” Ha! Coauthored work can sometimes also lead to disputes, such as the split between Yang Chen-Ning and Lee Tsung-Dao.

You get it, you get it; you don’t get it, you don’t get it

While pursuing his doctorate, Li asked a scholar he admired how to find an original idea. This was the answer. It is philosophical, but also true: either you have one or you do not. That simple.

You should not rely on others

This applies beyond academia. At a crucial moment, nobody is more dependable than yourself.

There is no standard process

You discover your own path. There is no need to be timid; trial and error is how you grow.

There is no shortcut

Keep your feet on the ground.

Every paper is not easy to publish

Persist.

How to Make Research Deeper and Wider

You must think deeply and uninterruptedly with concentration

Focus.

New ideas may come unexpectedly, but not surprisingly

When mountains and rivers seem to block every path, another village appears beyond the willows and flowers.

The first idea is the most difficult to obtain

The beginning is always the hardest part.

Usually more and deeper ideas will come when you write a paper

Progress builds on itself.

More papers, more ideas

If you are not good at it yet, practice more.

New Problems, old methods

Change, and what remains unchanged.

New research areas can be opened more easily with more experience

You do not necessarily have to chase every trend. Going deeper into your own field is also a good choice.

What It Takes to Publish Single-Author Papers

Ability: a challenge for senior scholars

There can be so many competing obligations that some scholars struggle to concentrate.

Courage: a challenge for young scholars

This is quite a challenge too.

Motivation: dreams + goals + conviction + pursuit + persistence + passion

Whether or not you ultimately achieve it, publishing a single-author paper ought to be an ultimate goal for every academic researcher!

Passion was a theme Professor Li returned to throughout the talk. Without it, continuing is difficult, let alone doing something well.

A Master

Over 36 years, from 1989 to 2025, he published 1,130 papers—an average of 31 a year. Over 35 years, from 1989 to 2024, 245 were single-author or first-author papers—an average of seven a year. Li offered an interesting comparison: even if we did not have to invent anything, could we take existing papers, reproduce their code, and typeset them in LaTeX at a rate of seven a year? Ha! A true master.

Time Management

For example, to arrive at Li’s lecture on time, I had to finish my Java exam in 25 minutes, then borrow a bicycle to get there. I had also overlooked one thing: arriving late might mean no seats. I realized something alarming—apart from this occasion, I had never really planned my time during university.

Role Models

These two people are Professor Li’s role models, and they should be ours too.

Academician Dong Yunmei

Dong supervised Li during his graduate studies at the Chinese Academy of Sciences. Li said there was only one place and nine applicants; he ranked first and beat the others. Ha! Dong also published single-author papers and deserves great respect.

Professor Kai Hwang

What impressed me most was that Hwang worked until he was 80, returned to China, and then taught at the Chinese University of Hong Kong for another five years. He independently authored more than ten books—a truly prolific writer. He also kept up with new developments, immediately studying and investigating new things. Lifelong learning.

Commitment from Beginning to End

Professor Li’s lecture was also full of affectionate stories about his wife. I will not elaborate, but he is quite the example of a loving partner… Whether in research or love, he has remained committed. He and his wife have also been remarkably successful in raising their children: all three attended Ivy League schools!

Health Comes First

There is an old analogy: health is the leading 1, and everything else is a zero after it. Without the 1, everything disappears. Health really matters. A reminder I keep giving myself…

Q&A

My first question was: learning conditions were nowhere near as good in your day, so how did you study? His answer was roughly: passion. On one occasion, he spent an entire day looking up material in an IBM library.

My second question was: what was the biggest difficulty you encountered, and how did you overcome it? He did not have an answer ready and said he would tell me later.

My third question was about his typical daily routine. It touched somewhat on his private life, but he still answered. He generally sleeps until he wakes naturally and does not stay up late. After getting up, he usually replies to WeChat messages—he has more than a thousand collaborators—including students’ questions and issues with papers. Sometimes he answers directly by voice call. After that, the priority is maintaining high efficiency.

Passion

Professor Li said that although he is 62, he feels no older than 18 inside. You really could not tell his age: his voice was resonant, he was passionate throughout a lecture lasting almost two hours, and he even had the same fashion-forward buzz cut as me. Honestly, although I am not far past 18, I have already lost much of the sharpness and passion of my younger self. I feel ashamed. I hope that fire returns one day.

Closing Thoughts

This was, in all, the most valuable lecture I have ever attended. Professor Li is humorous, engaging, and extraordinarily accomplished—a role model for my generation!