Carles Sáez

HILO DE TEOREMAS DE TOPOLOGÍA ALGEBRAICA Un calendario de adviento con teoremas de topología algebraica e hilos de comentarios de cada teorema

“It’s my honor to announce that the true pope is working with us from Avignon”

“Pope Leo was a math major? I wonder what his encyclicals will be like” Like this

@Isaaccarcacia @Juanandete Enhorabuena!! 🥳

Nuevo artículo, de matemáticas, publicado. degruyterbrill.com/document/doi/1…

Todo el mundo esperando la Jihad Butleriana y al final igual tenemos la Cruzada Butleriana

No es que yo sea nada religoso, pero PAPA BASADO: vatican.va/content/leo-xi…

Pope Leo just condemned the idea that "work dignifies man". We. Are. So. Back!

The Riemann Hypothesis, the Goldbach conjecture, the Twin Prime conjecture... They have all been extensively verified numerically. So for all practical purposes, these are "true" statements. If needed for an actual practical use, they could easily be verified even farther. The mathematical truth of these statements would only be "useful" to mathematicians that seek the truth and those who seek to understand *why* these statements are true. If an AI agent proved these statements but there were no mathematicians to understand and digest the proofs, what would be the point of such a proof? Even in the most optimistic of cases, where a super human mathematician agent exists that could prove or disprove (or declare undecideable) every statement, either there are human mathematicians that are there to understand the proof to explain it to other humans... Or there is simply no point for such a super human agent to exist in the first place.

I think this was lost in the noise of all the unit distance problem solve news! Paper from DeepMind: arxiv.org/abs/2605.22763…

@ibaa13_ Buen tema! Supongo que en el.tribunal habrá gente que sepa algo de eso, seguro que algo entienden

@csaez_math Descomposición de Hodge y la relación con la cohomologia de haces, desarrollos típicos de cualquier libro introductorio de geometría compleja

No sé cómo conseguir que mi tribunal entienda algo de mi tfg en solo 20 mins (va a ser el tfg más aburrido de toda la convocatoria)

getting into a rough period again and won’t be here for a while until i sort some things out. please in the meantime support my friends @AhmSuhad77381 @NewSandr2 please help @mobSlma1999 and his kids!! check out @ll4gaza_ofcl @sameerproject and @Handala_Relief please share ❤️

"Solve all open problems and finish all open research from this. Make no mistakes."

I suspect for the next few years, AI for math will be most useful for pushing back against our biases that (1) certain false conjectures are true, and (2) certain approachable problems are intractably hard. Mostly by trying things our judgment is now too clouded to try

Very interesting and wise commentary on AI for math

@willdepue @eigenknight I mean, they are basically "human-language simulators" and deep down all they do is next-token prediction, so it's hard for me to see how it could be otherwise with the current models

it’s kind of fucking ridiculous (and quite frightening) we‘re this far — the models are solving long standing problems in discrete geometry — yet the models do this still by thinking to themselves in plain english? that is easily interpretable? what the hell man

If you are a mathematician, then you may want to make sure you are sitting down before reading further.

@Daanniii6 Exactly! I remember doing my PhD thesis and thinking "finally! the reference with the result I was looking for!" Only to discover shortly after that it never appeared 🥲

@csaez_math it's especially bad because a lot of the time it's "In this paper, we lay the foundations so that the next paper to appear shows the exact result you've been searching for"...

A lot of discourse about AI hallucinated references, but in the 90s that was already a thing in gauge theory literature (and yes, most of those "to appear" never appeared)

This one is (IMO) pretty interesting! openai.com/index/model-di…