Nostr notes by Liu Yao 刘杳

Interesting. Strange that it's got an acronym as the title of ...

2026-10-12T02:41:00Z

In reply to nevent1q…dxms
_________________________

Interesting. Strange that it's got an acronym as the title of its wikipedia page. I asked AI for pronunciation, and apparently it's "seeks" or "seeks pee-oh-vee-em". Can't we find a better name for it?

I presume you don't have a strong aversion to feeding such ...

2026-09-22T07:23:44Z

In reply to nevent1q…at4y
_________________________

I presume you don't have a strong aversion to feeding such questions to an LLM. (No explanation of the same kind, if you don't mind a little spoiler.)

https://chatgpt.com/share/6ab22c08-5c10-83ea-85cd-4bf8da82a093

I wanted to see if AI can build an interactive visualization for ...

2026-08-20T06:37:00Z

In reply to nevent1q…4967
_________________________

I wanted to see if AI can build an interactive visualization for this. Still not happy with their 600-cell, but here's 120-cell: https://liuyao12.github.io/3-sphere/?mode=120-cell

seems perfect for AI to work out

2026-07-08T06:45:33Z

In reply to nevent1q…949f
_________________________

seems perfect for AI to work out

Another resource is lectures by Venkatesh ...

2026-06-15T15:09:57Z

In reply to nevent1q…wftm
_________________________

Another resource is lectures by Venkatesh https://www.youtube.com/watch?v=qMTUlD6SnIc

Sounds like translation surfaces but with "ridges"? ...

2026-06-12T03:32:37Z

In reply to nevent1q…fzx2
_________________________

Sounds like translation surfaces but with "ridges"?

https://en.wikipedia.org/wiki/Translation_surface

Will Sawin was one of the people who checked and distilled ...

2026-05-22T01:42:15Z

In reply to nevent1q…e85k
_________________________

Will Sawin was one of the people who checked and distilled OpenAI's proof. It's a safe bet that he had come to the explicit bound before the release, and chose to publish separately and simultaneously.

Presumbaly this does not cover quasicrystals/aperiodic ...

2026-04-27T05:40:51Z

In reply to nevent1q…vqxm
_________________________

Presumbaly this does not cover quasicrystals/aperiodic (mono-)tilings?

Appreciate your efforts! This has very much to do with ...

2026-04-27T03:40:47Z

In reply to nevent1q…fdsk
_________________________

Appreciate your efforts! This has very much to do with Conway's (or Thurston's) orbifold notation of space group that I hope you will get to :)

That's so much more clear than other accounts I've seen. ...

2026-04-25T06:44:20Z

In reply to nevent1q…ue28
_________________________

That's so much more clear than other accounts I've seen.

Just to make sure: in L -> G -> P (kind of funny you put 0 on one end and 1 on the other), L is understood as "translations", and P would be those fixing a point, hence "point group", right?

ArXiv already faces real challenges from AI generated papers, and ...

2026-03-14T13:20:06Z

In reply to nevent1q…hmp6
_________________________

ArXiv already faces real challenges from AI generated papers, and this doesn't seem to be the way to fence off what's coming.

I was just thinking the other day that anything you can prompt AI to generate within a few minutes would not be worth publishing. Existing papers should also undergo an evaluation by AIs (one AI reads it, summarizes the main ideas, and feeds to another AI to reconstruct.) And we should probably only archive the prompts.

Presumably sundials only work at specific lattitude? The Chinese ...

2026-02-25T01:58:40Z

In reply to nevent1q…9wnq
_________________________

Presumably sundials only work at specific lattitude? The Chinese ones can work anywhere, you just have to tilt it differently

How many books is that payment equivalent to, in terms of money ...

2025-10-31T05:35:21Z

In reply to nevent1q…x5st
_________________________

How many books is that payment equivalent to, in terms of money you actually would have received?

I can't speak for the authors, as I have never written any books. But I feel that the AI models trained on these books are still valuable, just that they can't make money out of them. If only there's a way to pay you each time a query (or the AI's response of it) references your books.

Does that means one should avoid travelling through the south ...

2025-10-16T02:08:23Z

In reply to nevent1q…7608
_________________________

Does that means one should avoid travelling through the south Atlantic? Not that it's on my future plans.

Most mathematicians who are drawn to science would naturally ...

2025-09-25T02:30:45Z

In reply to nevent1q…qkza
_________________________

Most mathematicians who are drawn to science would naturally favor physics, while chemistry seems like a mess (too many things to memorize, from high school experience). Perhaps crystollography would be an exception. Knowing your interests in entropy, I'd bet the "concept" of entropic bond would be your cup of tea: https://www.youtube.com/watch?v=q9_2UR1DsgA

There you have Bragg the Elder 😀

2025-09-05T12:22:41Z

In reply to nevent1q…6nk8
_________________________

There you have Bragg the Elder 😀

I vividly had this thought when I was like 10, an infinitely long ...

2025-08-05T22:52:39Z

In reply to nevent1q…sgs4
_________________________

I vividly had this thought when I was like 10, an infinitely long string of zeros, followed by a 1. I thought we all had to struggle with reconciling ourselves about this fact at some point (not necessarily having to involve definition/construction of reals), and apparently it's worse for grown-ups.

The best argument I heard, though I was mature enough when I heard it, is that "if they are truly different, tell me a number that's in between these two."

they've kept it alive from the days of Pythagoras, Euclid, ...

2025-06-30T06:34:16Z

In reply to nevent1q…x7qa
_________________________

they've kept it alive from the days of Pythagoras, Euclid, and Archimedes.

I remember this image on the door of Bill Thurston, in the last ...

2025-05-26T01:16:07Z

In reply to nevent1q…3vu6
_________________________

I remember this image on the door of Bill Thurston, in the last years of his life. Sharing w/ John Carlos Baez (npub1nf4…nqe4)

they've been doing great conservation work for a number of ...

2025-05-21T03:59:31Z

In reply to nevent1q…n92h
_________________________

they've been doing great conservation work for a number of years. They have a website https://homingleopards.org/

not the same sort of number (cardinal, not ordinal) ...

2025-02-28T07:13:08Z

In reply to nevent1q…aqqj
_________________________

not the same sort of number (cardinal, not ordinal) https://en.wikipedia.org/wiki/15_and_290_theorems

For theorem labels that appear in a paper then got stuck, I offer Zariski's Main Theorem

https://en.wikipedia.org/wiki/Zariski%27s_main_theorem

I don't know why people are not talking about their prior ...

2025-01-29T01:41:06Z

In reply to nevent1q…q80g
_________________________

I don't know why people are not talking about their prior work on math. Maybe they deliberately didn't mention that in their new reports?

Would be interesting to note that they are somewhat known in the ...

2025-01-26T02:04:48Z

In reply to nevent1q…kec6
_________________________

Would be interesting to note that they are somewhat known in the AI+proof assistant community, as the top winners of last year's AIMO competition had used their math model as the base model.

I had make contact with someone on their team, and got the impression that the company (with its background in hedge fund) had the vision to work on math related AI in the hope that it would pay off in their general model. I did not expect it to work, not to say so fast.

yes, I remember your posting about it on the other site.

2025-01-08T22:49:05Z

In reply to nevent1q…fjzq
_________________________

yes, I remember your posting about it on the other site.

Thanks for reminding me to get back on with a little project of ...

2025-01-08T07:52:40Z

In reply to nevent1q…5xjr
_________________________

Thanks for reminding me to get back on with a little project of mine, which lets you compute digits in your browser. Not there yet to do the AGM algorithm, but it occurs to me that you can design a whole course on calculus just on the various ways to compute pi. (I'm never a fan of the pop math side of it, but apart from the digits, pi is indeed awesome.)

https://mathstodon.xyz/@liuyao/113780063926562388

if i'm allowed to think out loud: it looks like the three ...

2024-12-18T02:11:45Z

In reply to nevent1q…5qpw
_________________________

if i'm allowed to think out loud: it looks like the three cynlinders are carefully positioned (touching on the outer surface), and it probably has to depend on the width of the Mobius band or you'd be able to widen it indefinitely. A self-intersecting band?

how did it know where to go? the article didn't give a clue.

2024-12-16T07:13:51Z

In reply to nevent1q…l7ym
_________________________

how did it know where to go? the article didn't give a clue.

I tried to use ChatGPT o1-preview to do some coding (for a ...

2024-10-26T09:01:05Z

In reply to nevent1q…h4k4
_________________________

I tried to use ChatGPT o1-preview to do some coding (for a different problem). Not too bad, but still requires a few back-and-forths. I'm hoping the full version of o1 is better at this kind of stuff. https://observablehq.com/@liuyao12/tiling-with-lattice-polygons

now they have no excuse not giving one to Ed Witten. He took ...

2024-10-09T03:33:52Z

In reply to nevent1q…a3da
_________________________

now they have no excuse not giving one to Ed Witten. He took ideas from physics and led to many great advances in mathematics!