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

Erdos 281 solved with ChatGPT 5.2 Pro

twitter.com

Post reply on HN
Print view View original

171–180 of 310 posts

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#171

Post by danielbln »

user3939382 wrote: post #33
mikert89 wrote: post #7

I have 15 years of software engineering experience across some top companies. I truly believe that ai will far surpass human beings at coding, and more broadly logic work. We are very close

They can only code to specification which is where even teams of humans get lost. Without much smarter architecture for AI (LLMs as is are a joke) that needle isn’t going to move.

Real HN comment right here. "LLMs are a joke" - maybe don't drink the anti-hype kool aid, you'll blind yourself to the capability space that's out there, even if it's not AGI or whatever.

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#173

Post by vercaemert »

charmpic wrote: post #168

I'm looking forward to chatgpt 5.3pro. I also use chatgpt 5.2pro for various program consultations. It's been very helpful.

I was hoping there'd be more discussion about the model itself. I find the last couple of generations of Pro models fascinating.

Personally, I've been applying them to hard OCR problems. Many varied languages concurrently, wildly varying page structure, and poor scan quality; my dataset has all of these things. The models take 30 minutes a page, but the accuracy is basically 100% (it'll still striggle with perfectly-placed bits of mold). The next best model (Google's flagship) rests closer to 80%.

I'll be VERY intrigued to see what the next 2, 5, 10 years does to the price of this level of model.

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#176

Post by supermatt »

klohto wrote: post #175
supermatt wrote: post #174

What does "solved with" mean? The author claims "I've solved", so did the author solve it or GPT?

When you use a calculator, did you really solve it or was it the calculator?

With a calculator I supply the arithmetic. It just executes it with no reasoning so im the solver. I can do the same with an LLM and still be the solver as long as it just follows my direction. Or I can give it a problem and let it reason and generate the arithmetic itself, in which case the LLM is effectively the solver. Thats why saying "I've solved X using only GPT" is ambiguous.

But thanks for the downvote in addition to your useless comment.

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#177

Post by BoredPositron »

glemion43 wrote: post #158
cubefox wrote: post #117

Earlier quoted context omitted.

This illustrates how unimportant this problem is. A prior solution did exist, but apparently nobody knew because people didn't really care about it. If progress can be had by simply searching for old solutions in the literature, then that's good evidence the supposed progress is imaginary. And this is not the first time this has happened with an Erdős problem. A lot of pure mathematics seems to consist in solving nea…

It shows that a 'llm' can now work on issues like this today and tomorrow it can do even more. Don't be so ignorant. A few years ago NO ONE could have come up with something so generic as an LLM which will help you to solve this kind of problems and also create text adventures and java code.

You can just wait and verify instead of the publishing, redacting cycles of the last year. It's embarrassing.

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#179

Post by _fizz_buzz_ »

xigoi wrote: post #160
Davidzheng wrote: post #135

Earlier quoted context omitted.

I actually don't think the reason is that they are easier than other open math problems. I think it's more that they are "elementary" in the sense that the problems usually don't require a huge amount of domain knowledge to state.

The Collatz conjecture can be stated using basic arithmetic, yet LLMs have not been able to solve it.

That is also one of the hardest problems.

Re: Erdos 281 solved with ChatGPT 5.2 Pro

#180

Post by jojobas »

cubefox wrote: post #117
xeeeeeeeeeeenu wrote: post #18

> no prior solutions found. This is no longer true, a prior solution has just been found[1], so the LLM proof has been moved to the Section 2 of Terence Tao's wiki[2]. [1] - https://www.erdosproblems.com/forum/thread/281#post-3325 [2] - https://github.com/teorth/erdosproblems/wiki/AI-contribution...

This illustrates how unimportant this problem is. A prior solution did exist, but apparently nobody knew because people didn't really care about it. If progress can be had by simply searching for old solutions in the literature, then that's good evidence the supposed progress is imaginary. And this is not the first time this has happened with an Erdős problem. A lot of pure mathematics seems to consist in solving nea…

It's hard to predict which maths result from 100 years ago surfaces in say quantum mechanics or cryptography.
Post reply on HN