>we uploaded SEVEN HUNDRED PAPERS TO GUTHIB!!11!>look inside>its slop work pieced together by the equivalent of gpt plus models>didnt even bother having multiple agents work for days on each paper one paper at a time
>>17070457If it's slop how come humans could not come up with it?
Yep they’re just spamming sloppa faster than anyone can actually proof read it and validate/check it actually is what they’re claiming
>>17070469you don't know what you're talking and about
>>17070459its precisely because its slop that humans would never publish it(or at the very least nobody would pay attention to it)
>>17070470Whatever you say, Rajeeb.
>>17070472You're out of your depth retard, there are thousands of terse and nearly unreadable papers out there, you will come across those as well once you actually start researching and stop shitposting on 4chan
>>17070469Look at this retard kek
>>17070472>"riemann zeta has no zeros with real part above 7/8" is a slop result no human would pay attention tocome on now
>>17070472cringe
jeet squad has arrived
>>17070472>humans would never publish itAnon, do you understand that many of these results are not just completions of on-going research? There are tons of results in that repo that are unique and no mathematician has come out so far is "stolen" from them.So my question stands: how come humans could not arrive at the results if it's slop?
>>17070469>>17070457>>17070472>t. undergrad that spends his free time seething over technology instead of studyingngmi, ted didn't make it either
>>17070485>absence of evidence is evidence of absence>>17070480>holy shit it bruteformed and found le curazy exception you could probably find with traditional software>>17070476were talking about content not prose
>>17070494it didn't bruteforce shit, it was a positive proof by contradiction, not a counterexample to anything
who cares if it's slop as long as there's a reasonably understandable lean statement? If someone cares about understanding it in detail or trying to solve it by hand without looking at the AI solution nobody's stopping them.Anyway why haven't they formalized IUTeich yet and checked the abc proof?
>>17070498>an_iq_too_high?.jpg
>>17070494>were talking about content not proselmao fucking retard. Terse papers aren't terse due to their prose, it's because of the content: lack of examples, lack of exposition, etc. the prose of math papers is relatively standardized
>>17070503do you know what a proof by contradiction is?
>>17070494Let me relay you back to my original question >>17070485>how come humans could not arrive at the results if it's slop?
>>17070504sure but that's the slop that wont be read(if you recall my original post)>>17070506its an exception(to the contradiction)mindblown.png
>>17070509it will be read, there are entire areas with widespread terrible standards for papers.stop larping and acting like you know anything about the field lmfao
>>17070508is anyone voraciously gobbling up the github and looking for them? probably not
>>17070512Fuck me you people are retarded. I bet some lone mathematician just uploaded the quasi RH result to GPT a month ago as well, all of these areas were just on the brink of making all of these discoveries all at the same time and OAI scooped all of them.Use your brain.Also like how the convo shifted from "b-but it's slop!!!" to coping over the results being real by claiming they're all stolen. Thank you for your concession
>>17070512nta but even if that xitter schizo were right, that's entirely on the researchers. Imagine uploading draft papers to a website that publicly admits to training on everyone's conversations LOL. I consider anything I upload to these websites to basically be public
>>17070517>slop slop slop slop>specific notation(stolen)>slop slop>slop result(unfinished)(probably wrong)woah.... geepeetee SOLVED math
>>17070524would you like to bet money on the quasi-RH result being fake or stolen?
>>17070524>(probably wrong)You're quite literally the only person that is saying that. No mathematician is denying these results, they're mad over bad etiquette and no research artifacts beyond the compressed result papers.
>>17070457>>17070503>>17070509>!!!111>.png>.jpgTalk like a normal human being for once you obnoxious faggot. /r9k/ has been a curse on this website and the Internet in general, my god.
>>17070532OP is either a bot or simply suffering from AI psychosis. He has been spamming the board for months.
>>17070457White coded mentally ill histrionics.
>>17070530I think like two of them have been found to be wrong so far, but even if 30% of them were wrong, a machine that generates correct novel mathematical results in 3 hours of computation (on average) 70% of the time is still revolutionary and OP is a faggot
>>17070533Either a tourist that got here from twitter greentext reposts or an LLM trained on /r9k/ slang.Either way, without a doubt the worst kind of poster. Every single post is devoid of any substance and just screams "GIVE ME ATTENTION," this is even worse than the Atheism vs. Religion spam from 15 years ago or the /pol/ anti-vax schizo newfags during Covid
>>17070501There was a formalization project that eventually got stuck at the same point where Scholze-Stix objected https://ncatlab.org/nlab/files/LANAProject-Report-July2026.pdfI think it's pretty clear now that if Mochizuki truly has an argument, it's not yet well formulated enough. He basically keeps on calling people in the community retarded for not filling in these "obvious" details yet isn't doing that himself.Could be interesting to see this new internal model attempt IUTT though lol
>>17070544ive been here forever since pre orange man i just got bored
“Mathematicians did not ask for this work to be done…”I didn’t know we had to ask.Should I stop calculating my daily calorie intake until I get express permission from mathematicians?
>>17070538Yeah and most math faculties have received free GPT Pro subscriptions by now, they can see its capabilities for themselves. The speed at which these results are produced is most incredible about it, even my advisor has told me how Astra Pro solved a smaller but nevertheless real problem in their research in 80 minutes, which they've been stuck on for the past 2 years (certainly not working full-time on it nor as significant as any of the ones OpenAI's internal model solved but it was a remaining problem). He said he would've probably expected it to be solved in the next 3-5 years by a young motivated researcher.I genuinely have yet to meet these "mathematicians" calling for AI to be banned IRL, most people I come across just seem excited and worried about it, but a healthy and reasonable kind of worry, none of that insane shit like OP or the retards over on reddit flat-out banning any discussion on it because it's AI and calling for the crucifixion of mathematicians that dare to work for an AI company
>>17070457That is not written by Tao.
>>17070557nta but you aren't "just bored" when you spend months shitting up /sci/, you're addicted to shitposting
>>17070570>spend monthsyou dont know me dumbass im barely on /sci/ these days im just here today because /g/ banned me
>>17070578Idk man, I just came here after the announcement and it seems an anon immediately recognized your schizo ass. I trust resident anons on /sci/ enough to call out the resident schizo, hate to say it
>>17070459humans came up with all of itAI just filled in some gaps to make slop proofs that quite literally nobody will ever read
>>17070580were bound to all sound the same since the culture has been relatively unchanged for the past decade or so. wojaks and pepe should have died off by now
>>17070529I wouldn't bet money but I view the quasi-RH result with just as much skepticism as I would if any other high-level mathematician announced it. Especially now that they've withdrawn Hodge.
>>17070584They are reading it though, or are you claiming it's AI that found the sign error invalidating 3 papers?
>>17070588>just as much skepticismI kind of doubt that. I don't think most mathematicians doubted the results when the proof of unramified GLC first dropped, everyone knows full well that it takes months if not years to review but there's generally no widespread doubt over its validity.
>>17070598You're right, I should've said significantly more skepticism
>>17070557>>17070578>>17070585You're not fooling anyone, your writing style is very distinctive, be it the cringe /r9k/ slang, forgetting to press space before a parenthesis, lack of reading comprehension, etc. it's trivially easy to just look for these markers in the archive and find HUNDREDS of posts
>>17070494Confirmed that it's actually the naysayers that have absolutely no clue what is in the release contents.
>>17070712Well what can I say believe whatever you want but its not me nigga
>>17070712>forgetting to press space before a parenthesisstylistic choice i see it used outside of 4chan all the time>lack of reading comprehensionwhatever you say buddy>cringe /r9k/ slangits very common across boards especially baitposters. cmon now
>>17070469>>17070457dumb miggers
>>17070517The Riemann hypothesis for curves over finite fields has been proved by human, so it's not like no human ever proved a restricted version of Riemann Hypothesis
>>17070469We will look back on these days like we look back on the NFT/block chain era
>>17070459Human mathematicians generally have enough self awareness to not post such shamefully sloppy articles while claiming they are actual solutions. Though, there are apparently enough humans who misunderstand LEAN to the point that they assume that an accompanying compiled LEAN script (which they didn't read) returning "valid" automatically implies the mathematics (which they also didn't read) are also valid. It's going to be the same story as it was with Navier-Stokes. The LEAN doesn't match the math, and the math doesn't match the LEAN.
STOPTHEMATHCOUNTTRUST THE EXPERTS!
>>17070457Cope
>>17070485>how come humans could not arrive at the results if it's slop?I could have but I didn't want to
>>17070498>proof by contradictionYawnYou never learn anything from proofs of this nature
>>17070469Make Millennium million dollar AI slop and have them to come up with the proof in checking it. Genius idea. What do we know what they think is correct or not MISTER!
>>17070498And what is the easiest way to get a proof by contradiction.
>>17071510assume a counterexample exist
>>17071510Assume a spherical cow
>>17070457Well then I guess mathies have nothing to worry about. They can stop shitting their diapers now because the AI is incompetent and won't replace them.
MATHIE ARE YOU OKARE YOU OK MATHIE
In the future, how will mathies prove that any problems they solve were indeed solved by them and that they didn't use AI?
>>17071584Even as of now mathematicians are (hearvily?) supported by AI
>>17071584there will be no such thing as professional mathematicians, math will become just a hobby. In the ancient and medival times, almost all math discoveries were made without any attribution.
>>17071584In the future there will be secret guild of mathematicians, where all the techniques, methods and partial progresses are highly guarded secrets and final results are released as a collective.
>>17071589wild its medieval times
>>17071589
>>17071646More like pic related
>>17070591It was a guy with Astra
>>17070483They're especially worried since AI is making the entire Indian Subcontinent obsolete
>>17071575Are you?
Is there a thread about this topic that isn't a massive cope session for the biggest high school math pseudo-intellectuals?Every guy who majored in maths will tell you that math is bitch work. If AI can cut down on this work, it's an improvement.
>>17071711Same shit with programming. Only the most autistic were retarded enough to claim programming was as anything but monkey work. Now that it's gone, we're all better off.
>>17071711Mostly not because normal people don't start threads about AI garbage and the schizo spams all of them anyways.
>>17070457>Mathematicians did not ask for this work to be done.Cry me a river.
>>17071281Everyone learns this lesson at some point in their life. For me it was hoarding refined metal on TF2 thinking keys were never going to take off.
>>17070457CLEAN IT UP MATHIE:>Whenever I referee a paper, or receive a thesis or project from a student, I apply the same standards that most of us do. If a paper or thesis is so badly written that I cannot get past the first page without considerable effort, I reject it and ask for a new, readable version.>With these AI-generated papers, I feel that we, as a community, are not applying the same standards of quality and rigour. They can simply release multiple 100+ page papers claiming to have solved this or that problem, often with redundant arguments, unclear logical structure, multiple dead ends, and strange or unsettling terminology — in other words, slop. And we are then expected to go through it, check it, clean it up, simplify it, and explain what is actually going on.>This comes at a considerable cost to us in terms of time and effort, while they can simply move on and slop-bulldoze the next conjecture. And, of course, the credit remains theirs. In some sense, we are willingly contributing to our own demise.
>>17070457AI trannies BTFO
>>17072009>OpenAI just wanted to test its internal model on some hard math, and it turned out that anything less than the hardest open problems no longer suffices for this purpose. This is why all non-Millennium-Prize solutions were reached in a single-agent setting after 3 hours of thinking: OpenAI was literally just testing its model on these questions, similarly to how I run prinzbench.>Once OpenAI had the solutions (which were a side effect of sorts of benchmarking its model), a question arose as to what exactly should be done with these solutions, which are genuine advances in the field of mathematics. OpenAI - correctly - decided to make them public.>It seems clear that many other problems in the set of ~4,000 are very likely solvable by this model with more compute than just ~3 hours of Pro-level thinking. Whenever a more powerful checkpoint of this model is ready to be tested, OpenAI will likely have even more to share.>The main reason these papers exist is that OpenAI used unsolved math problems to evaluate their internal models for the purpose of improving their capabilities. They had to use hard open problems because these are the only ones still posing a challenge to frontier AI models. I believe that, from OpenAI’s point of view, the papers are essentially a byproduct of their internal development. I believe that, from OpenAI’s point of view, the papers are essentially a byproduct of their internal development.>It’s clear many papers were never reviewed by a human: for instance, I took their example #254 (an Artin group with no CAT(0) action): their group has 116 generators, but after feeding it to Astra, in 10 minutes I got a new example with 12 generators.
>>17072026tl;dr these are just traces from an AI's training and testing, not serious endeavors to do math research
>>17072026So it means AI can only solve human posed problems, but cannot discover new theorems that human cannot even imagine on its own?
Jesus christ, he's retarded.
>>17072032LLMs fundamentally cannot create. Neural networks are all pattern recognition.
>>17070457Terence is retarded.
>>17070480We never cared about the Riemann hypothesis anyway.
>>17071281>We will look back on these days like we look back on the NFT/block chain erapromise anon , like allegedly people have been using this for programing for example , even if it was just stolen google searches , that would still be impresive.is that another lie?
>>17071821I love how they still believe they are necessary for this when we will just make an AI that will clean up all the pages into a human readable format in a day. They are less relevant than their worst janitorial nightmares
>>17072069>Neural networks are all pattern recognition.You mean like the human brain?
>>17071281>We will look back on these days like we look back on the NFT/block chain eraLol lmao evenThe next 10 years are going to be hard on you
>>17070457didn't he troon his son?
>>17070457someone tell TT that math aint science hes retarded
>>17072343LLMs are not a human brain.
>>17070457>Mathematicians did not ask for this work to be done>Math must remain ethically researchedImagine thinking you own a discipline.Imagine claiming there is ethics in math of all things.Can't wait until based /g/ chads completely owns these whining autististic pseuds with their machine waifus.
>>17072032I have no fucking idea how you managed to misread that from what I posted.
>>17070512kek AI is basically the electric jew
>>17070508>how could human not do prime numbers to the millionth digits if it's slop
>>17072767when the university is an apparatus for control then expect "ethics" in every discipline that the university recognizes
>>17072767proof must be ethical. to be ethical, it must be done by me
>>17070512Any mathematicians who worked on those problems are looking at the proofs, yes.
>>17072331who'se gonna pay for it?
>>17072063That's what happens when you put all your points in math.
>>17072026>We just wanted to test an internal model>by not checking the work was correct
it's just altman and thiel spamming up the web about their products to ger more money from boomers
