They're already in university, and the primary school system is only just starting out?
Chapter 124 The Draft Written by God
Chapter 125 The Draft Written by God
However, this openness to cutting-edge fields has also brought troubles to editors like Bob.
On the one hand, the underlying structure of artificial intelligence is neural networks, which has led countless ambitious people to regard this direction as fertile ground for "watered-down papers".
They often just make minor, insignificant modifications to a few classic neural network structures, rearrange them haphazardly, add a seemingly profound but actually nonsensical mathematical proof, and finally run it on a few overused introductory datasets. A freshly published, hot-buttoned, low-quality paper is then ready.
This situation has intensified, especially after the explosive popularity of large model technology. JAMS's submission inbox is flooded with this kind of "academic garbage" almost every day.
On the other hand, there are indeed top scholars with extremely profound mathematical foundations who, at the level of mathematical theory, have proposed truly groundbreaking improvements or even entirely new algorithmic frameworks for artificial intelligence algorithms.
However, these kinds of papers often use a lot of difficult, obscure, or even very trendy mathematical tools and methods.
Looking at the whole world, you can count on your fingers the number of people who are qualified and capable of reviewing manuscripts of this level.
These mathematical giants are all incredibly busy people, each with their own set of responsibilities.
Even for someone as known for being easy to talk to as Terence Tao, getting him to review a manuscript still depends on luck.
Like now, if someone arrives a day late, they might have to go on a business trip for a month. Bob's current predicament is the norm.
The problem is, besides Terence Tao, Bob couldn't think of a second person for the time being.
The key point is that the paper he has is fundamentally different from those poorly made, low-quality papers!
This point was unanimously confirmed by the most senior editors in the JAMS editorial department during the initial review stage of the paper.
Although the group of them spent most of the day studying the paper together, they could only barely understand a small part of it. But that was enough for them to conclude that the author of the paper possessed an extremely terrifying level of mathematical ability!
That's why the editor-in-chief specifically instructed Bob to make sure the peer review of this paper was completed before the next issue was published.
So Bob had no choice but to bite the bullet and continue making a final effort.
"Professor Tao, I assure God, this paper will definitely pique your interest!" Bob's tone was pleading.
He was absolutely not exaggerating in the slightest.
Although the main body of this paper was basically like reading gibberish to him given his mathematical level.
However, he studied the abstract of the paper word by word, very carefully!
In the abstract, the author states clearly that he will provide a completely new solution to the "generative large model illusion problem" that has baffled both academia and industry!
Bob, seeing Tao Zhexuan's still somewhat hesitant face, decided to add fuel to the fire.
"The main framework used by the author in the paper is differentiable manifolds. Professor Tao, in this field, besides Professor Shing-Tung Yau, who is currently in China, you are the most qualified to review it!"
Bob's tone carried just the right amount of flattery, "Considering how many years we've worked with JAMS, could you just take a look, even just skim through the whole thing, please?"
Upon hearing this, Terence Tao let out a long sigh.
To be honest, if it were any other time, after hearing Bob put it this way, he would definitely take the time to check it out.
But today, it really won't work.
Starting in the first half of this year, the top leadership of the United States cut research funding across the entire University of California system without warning, and his laboratory was not spared either.
The main reason I accepted Jasker's invitation to fly to Silicon Valley to help him test the mathematical capabilities of his new model was that Jasker promised to provide his lab with a large sum of generous funding over the next three years.
This money is a godsend for Terence Tao right now.
However, Bob seemed to read his mind and immediately seized the opportunity to further persuade him: "Professor Tao, I know you're going to see Jasker. But think about it, what if—"
I mean, what if, by some chance, the paper I brought today contains some insightful ideas that could help with the training and development of large models? That might even foster a deeper collaboration between you and Mr. Jasker!
Upon hearing this, a flicker of interest finally appeared in Terence Tao's deep eyes, hidden behind his gold-rimmed glasses.
Although in the past two years, nine out of ten papers in the so-called "artificial intelligence" field submitted by JAMS and other journals have been a waste of his time, he had even subtly expressed to the editorial department that he hoped they could be more rigorous in the preliminary review stage.
But today, seeing Bob's unprecedentedly resolute attitude, and recalling his words just now—
Terence Tao was indeed somewhat tempted.
He glanced at the time on his phone, pondered for a moment, and then made up his mind.
He first walked to the side, made a call to his assistant, and succinctly instructed, "Change my flight, at least postpone it by an hour."
Then, he hung up the phone, turned around, and extended his hand to Bob, who looked on with surprise.
"Give it to me."
"Okay, okay!"
Bob felt like he'd been granted a pardon. He quickly opened his briefcase, carefully took out a printed copy that was meticulously protected in a folder, and handed it over with both hands.
Tao Zhexuan took the printed manuscript, his eyes first falling on the title.
【On the Boundary of Differential Manifolds in Generative Models and the
Control of Hallucinations
Differential manifold boundaries and illusion control in stem-generating models
The title alone caused a barely perceptible hint of approval to flash in Terence Tao's deep eyes, hidden behind his gold-rimmed glasses.
"Differential manifold..." he murmured to himself.
Approaching the problem from this angle, and attempting to solve the "illusion" problem that is prevalent in large models, is indeed a theoretically feasible and highly ambitious technical path.
In fact, he himself had seriously considered the possibility of this direction more than once in the past two years.
But each time, he only scratched the surface and never really went into depth.
The reason is simple: this road is just too difficult to travel.
The mathematical tools and theoretical methods required are not only numerous, but also span several vastly different fields, including differential geometry, algebraic topology, stochastic processes, and optimal control.
Just by thinking about it, Terence Tao can conjure up at least a few key steps that require rigorous proof and massive calculations.
The sheer volume of work was so immense that even he felt somewhat daunted by it.
It's not that he felt he didn't have the ability to complete it.
Rather, he simply had too many ongoing projects and research topics, each of which consumed a great deal of his energy, leaving him with no spare capacity to start a new one.
That's why, when he saw that the author of this paper actually dared to take on this challenge, Terence Tao felt a surge of goodwill towards this complete stranger.
He continued looking down, his gaze landing on the author's signature.
Lin Chen
Chen Lin?
Terence Tao raised an eyebrow slightly.
Chinese? Or of Chinese descent?
He was 100% certain that he had never heard of this name in any academic journal or conference.
However, he did not show any contempt because of this.
After all, Asia has never lacked outstanding mathematicians.
Moreover, just look at the uniformly Asian faces in the US National Mathematical Olympiad teams in recent years, and then recall the ridiculously high proportion of Chinese engineers in the top tech giants I've visited...
Terence Tao believes that even if the author was just an unknown young man before, it is perfectly reasonable for him to produce research results of considerable value.
He composed himself and focused all his attention on the main body of the paper.
Several minutes passed...
Ten minutes have passed...
Tao Zhexuan's body remained motionless, as if frozen in place, maintaining the initial standing posture.
His deep eyes were fixed on the printed manuscript, his brows furrowing deeper and deeper.
Seeing Terence Tao's unprecedentedly focused expression, Bob, standing nearby, subconsciously slowed his breathing, afraid of making the slightest sound and disturbing the legendary scholar's thoughts.
The little expectation in his heart was now expanding rapidly, like an inflated balloon.
Five minutes later, Terence Tao finally made a move.
He casually leaned the suitcase, its handle already extended, against the wall, then silently picked up the few A4 sheets of paper.
She turned and walked into her study, closing the door behind her with a "click."
"call...
'
Seeing this, Bob finally breathed a long sigh of relief and relaxed completely.
He stretched his shoulders, which were stiff from being tense for so long, and casually sat down on the sofa in the living room, a smug smile unconsciously creeping onto his lips.
Having worked with Terence Tao for so many years, he knows the professor's style of doing things all too well.
If he chooses to take the paper back to his study.
There's only one explanation—the content of this paper has caught his attention!
He needs a quiet environment to carefully and rigorously verify the arguments and calculations within!
Inside the study, Terence Tao laid the printed manuscript flat on his large desk and took out a stack of snow-white draft paper and a pen from the drawer.
Without the slightest hesitation, he began to verify the calculations line by line, starting directly from the first part of the paper.
"How to efficiently solve the high-dimensional Monge-Ampere equation numerically to calculate the boundary of the data manifold in a large-scale generative model."
""
How does the geometry (such as curvature) of a singular set quantitatively affect the probability of hallucinations?
"Design a calibrator based on manifold boundary detection" to verify and correct the generated content.
"The author's line of reasoning is frighteningly clear, and the logical chain is airtight."
The mathematical tools and proof methods he used were novel and ingenious, and combined seamlessly, so much so that Terence Tao could not find a single flaw in his verification process.
He couldn't even find a single area where he could make improvements.
Aside from a few obvious instances of awkwardness from AI-assisted translation in the writing style, the core mathematical calculations and proofs...
The mathematical part of this paper is perfect.
Terence Tao's first reaction was to deny his own conclusion.
As a mathematician, he prefers to believe that there must be a very clever logical trap hidden in this seemingly flawless argument, one that is enough to fool most reviewers!
Can you fool my eyes?
interesting.
A subtle smile curved his lips, and a fierce fighting spirit, as if he had met his match, ignited in his deep eyes.
Time passed by, second by second.
An hour passed quietly.
When Terence Tao's pen made its final stroke on the draft paper, he fell silent.
He looked at the printed manuscript of his paper, which was still on the first page, and then at the several sheets of draft paper next to him that were filled with calculations. After a long while, he slowly leaned back in his chair and let out a long sigh.
That breath carried a hint of frustration and... that he himself was unaware of.
...shocking.
"Amazing..."
""
He uttered the word softly, his voice filled with indescribable complex emotions.
If an hour ago he was still mostly skeptical, firmly believing that he could find the hidden critical error in this perfect argument.
Now, after an hour of continuous, intensive verification and reverse reasoning, he was almost certain that the argumentation process of this paper was indeed without any problems.
It's so perfect, it's like a draft written by God himself.
Besides "Amazing," he couldn't think of a more suitable word to describe his feelings at that moment.
If possible, he really wanted to meet the author of the paper right away and have a good talk with him.
However, Terence Tao also knew that there wouldn't be many opportunities in the near future. For the next whole month, he would be extremely busy in Silicon Valley at Jason Musk's AI company.
In the living room, Bob, utterly bored, had taken his back-and-forth walking exercise to the extreme.
Finally, he turned his attention to the glass fish tank on the entryway cabinet, and began to tap the glass of the tank with his finger, teasing the leisurely goldfish inside, in order to pass the time during the agonizing wait.
Just then, the heavy solid wood door to the study was pushed open from the inside with a "click".
Wearing casual clothes, Terence Tao walked out of the room with the printed copy of his thesis in his hand.
"Professor Tao?"
Seeing this, Bob perked up and quickly went to meet him, his tone a mix of nervousness and anticipation: "How did it go?"
Terence Tao did not answer him directly.
He walked straight to the wall, picked up his suitcase, carefully stuffed the printout into a hidden compartment, and then said without looking up, "I need some time. I'll give you an answer by the day after tomorrow at the latest."
Upon hearing this, Bob's breath caught in his throat for a second.
Immediately afterwards, an uncontrollable surge of joy rushed from the bottom of his heart to his brain!
It's done! It really is done!
Having collaborated with him for so many years, he knew this professor all too well!
If he doesn't return a paper to me with a smile after reading it the first time, that's enough to show that the paper's content has successfully captured his attention!
What's two or three days?
For a heavyweight paper that may have a profound impact on the entire industry, no review process is too rigorous!
The lengthy and repeated scrutiny and verification is not only the rigor expected of a top mathematician, but also the minimum respect a scholar should show to the field of his research!
Bob could almost foresee the uproar that the next issue of JAMS would cause in the academic and even scientific communities because of the publication of this paper!
In the world-renowned field of artificial intelligence, the illusion problem that has plagued all practitioners for a long time may have a clear solution!
JAMS will be the pioneer that sets up a signpost at the entrance to this road and draws everyone's attention.
This is not only a huge boost to the academic value of JAMS!
This is the best reward for editors who have consistently accepted innovative and cutting-edge manuscripts over the years!
>
You'll Also Like
-
My stats are higher than yours!
Chapter 177 2 hours ago -
Naruto: Random Flying Thunder God Technique? Intercepting the Dark-Dark Fruit
Chapter 168 2 hours ago -
Fairy Tail: Capturing the Demon Mirajane at the Start
Chapter 107 2 hours ago -
For horse racers, making money is victory!
Chapter 195 2 hours ago -
One Piece: The story was spoiled for Im Saint right from the start.
Chapter 211 2 hours ago -
If you dress up as a cute boy, can you do whatever you want just by acting cute?
Chapter 187 2 hours ago -
I became an immortal on earth.
Chapter 481 2 hours ago -
Cross through Teyvat, and you'll become stronger by taking the hits!
Chapter 426 2 hours ago -
Renting a young, attractive woman to take home for the Chinese New Year infuriates the entire villag
Chapter 317 2 hours ago -
Wasn't he supposed to be beaten to death? Why did this powerful cousin love him so much?
Chapter 3000 2 hours ago