Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

OpenAI claims gold-medal performance at IMO 2025

twitter.com

Post reply on HN
Print view View original

581–590 of 737 posts

Re: OpenAI claims gold-medal performance at IMO 2025

#581

Post by peripitea »

Rioghasarig wrote: post #436
kenjackson wrote: post #155

Earlier quoted context omitted.

Depends. I’ve been using it for some of my workflows and I’d say it is more like a solid junior developer with weird quirks where it makes stupid mistakes and other times behaves as a 30 year SME vet.

I really doubt it's like a "solid junior developer". If it could do the work of a solid junior developer it would be making programming projects 10-100x faster because it can do things several times faster than a person can. Maybe it can write solid code for certain tasks but that's not the same thing as being a junior developer.

It can be 10-100x faster for some tasks already. I've had it build prototypes in minutes that would have taken me a few hours to cobble together, especially in domains and using libraries I don't have experience with.

Re: OpenAI claims gold-medal performance at IMO 2025

#582

Post by yahoozoo »

famouswaffles wrote: post #491

From Noam Brown https://x.com/polynoamial/status/1946478258968531288 "When you work at a frontier lab, you usually know where frontier capabilities are months before anyone else. But this result is brand new, using recently developed techniques. It was a surprise even to many researchers at OpenAI. Today, everyone gets to see where the frontier is." and "This was a small team effort led by @alexwei_ . He took a resea…

That brand new technique? Training on the test data. /s

Re: OpenAI claims gold-medal performance at IMO 2025

#583

Post by charlieyu1 »

CamperBob2 wrote: post #558
charlieyu1 wrote: post #487

Earlier quoted context omitted.

Only have basic o3 to try. Spent like 10 minutes but did not return any response due to a network error. Checking the thoughts, the model was doing a lot of brute forcing up to n=8, and found k=0,1,3, but no mathematical reasoning was seen.

See how this compares to what you got from o3: https://chatgpt.com/share/687bf8bf-c1b0-800b-b316-ca7dd9b009... It convincingly argues that Gemini's answer was wrong, and Gemini agrees ( https://g.co/gemini/share/aa26fb1a4344 ). So that's pretty cool, IMO. Pitting these two models against each other in a cage match is an underused hack in my experience. Another observation worth making is that (looking at the Github l…

This answer from o3 looks better. There are still some holes, Lemma 1 works on the four right-most columns, then the model tries to apply it to any four columns. "Hence at least four columns lack a vertical line; take the last four columns above." is a slip in logic, and Lemma 1 doesn't work for any two columns. For example, if we choose columns 1, 3, 5, 7 as columns lacking a vertical line, then we can take a sunny line with slope -1/2 and this will meets at all these four columns. Still, this looks more promising than what I got.

Interesting result from Gemini, I don't know its thought process but it seemed like Gemini tried to improve from its own previous answer and then got there.

Re: OpenAI claims gold-medal performance at IMO 2025

#584

Post by tmule »

archon1410 wrote: post #477
tmule wrote: post #433

Earlier quoted context omitted.

Discussions about Indian politics or the Indian psyche—especially when laced with Indic supremacist undertones—are off-topic and an annoyance here. Please consider sharing these views in a forum focused on Indian affairs, where they’re more likely to find the traction they deserve.

It is not "supremacist" to believe that depriving hundreds of millions of people from higher education in their native language is deeply unjust. This reflection was prompted by a comment on why Indian languages are not represented in international competitions, which was prompted by a comment on the competition being available in many languages. Discussions online have a tendency to go off into tangents like this. I…

> self-loathing elites in India

Your disdain for English-speaking Indian elites (pejoratively referred to as ‘Macaulayites’ by Modi’s supporters) is quite telling. That said, as I mentioned earlier, this kind of discourse doesn’t belong here.

Re: OpenAI claims gold-medal performance at IMO 2025

#585

Post by davis »

demirbey05 wrote: post #4

Progress is astounding. Recently report published about evaluation of LLMs on IMO 2025. o3 high didn't even get bronze. https://matharena.ai/imo/ Waiting for Terry Tao's thoughts, but these kind of things are good use of AI. We need to make science progress faster rather than disrupting our economy without being ready.

Here they are: https://mathstodon.xyz/@tao/114881419368778558

Re: OpenAI claims gold-medal performance at IMO 2025

#587

Post by simonw »

cycomanic wrote: post #574
logicchains wrote: post #533

Earlier quoted context omitted.

I guess your work doesn't involve any maths then, because then you'd see they're capable of solving maths problems that require a non-trivial amount of reasoning steps.

Just the other day I needed to code some interlocked indices. It wasn't particularly hard but I didn't want to context switch and think so instead I asked gpt 4o. After a back and worth for 4 or 5 times, where it gave wrong answers I finally decided to just take a pen and paper and do it by hand. I have a hard time believing that these models are reasoning, because if they are they are very poor at it.

GPT-4o isn't classified as a "reasoning" model (by common 2025 terminology at least) - I suggest trying again with o3 or Claude 4 or Gemini 2.5.

Re: OpenAI claims gold-medal performance at IMO 2025

#589

Post by krackers »

torginus wrote: post #580
rafael859 wrote: post #185

Interesting that the proofs seem to use a limited vocabulary: https://github.com/aw31/openai-imo-2025-proofs/blob/main/pro... Why waste time say lot word when few word do trick :) Also worth pointing out that Alex Wei is himself a gold medalist at IOI.

In transformers generating each token takes the same amount of time, regardless of how much meaning it carries. By cutting out the filler from the text, you get a huge speedup.

Except generating more tokens also effectively extends the computational power beyond the depth of the circuit, which is why chain of thought works in the first place. Even sampling only dummy tokens that don't convey anything still provides more computational power.

Re: OpenAI claims gold-medal performance at IMO 2025

#590

Post by yorwba »

gafferongames wrote: post #59
Invictus0 wrote: post #58

Earlier quoted context omitted.

The "AI" that won Go was Monte Carlo tree search on a neural net "memory" of the outcome of millions of previous games; this is a LLM solving open ended problems. The tasks are hardly even comparable.

And then they created AlphaGo Zero, which is not trained on any previous games, and it was even stronger! https://deepmind.google/discover/blog/alphago-zero-starting-...

AlphaGo Zero was also trained on millions of games, they just weren't games played by human players.
Post reply on HN