The LLMs that take 10 attempts to un-zero-width a , telling me that every single change totally fixed the problem, are cracking the hardest math problems again.
Erdos 281 solved with ChatGPT 5.2 Pro
141–150 of 310 posts
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#142Earlier quoted context omitted.
[flagged]
I firmly believe @threethirtytwo’s reply was not produced by an LLM
regardless of if this text was written by an LLM or a human, it is still slop,with a human behind it just trying to wind people up . If there is a valid point to be made , it should be made, briefly.
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#143Earlier quoted context omitted.
The model has multiple layers of mechanisms to prevent carbon copy output of the training data.
forgive the skepticism, but this translates directly to "we asked the model pretty please not to do it in the system prompt"
That might be somewhat ungenerous unless you have more detail to provide.
I know that at least some LLM products explicitly check output for similarity to training data to prevent direct reproduction.
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#144Earlier 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…
There is still enormous value in cleaning up the long tail of somewhat important stuff. One of the great benefits of Claude Code to me is that smaller issues no longer rot in backlogs, but can be at least attempted immediately.
The difference is that Claude Code actually solves practical problems, but pure (as opposed to applied) mathematics doesn't. Moreover, a lot of pure mathematics seems to be not just useless, but also without intrinsic epistemic value, unlike science. See https://news.ycombinator.com/item?id=46510353
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#145Earlier quoted context omitted.
The model has multiple layers of mechanisms to prevent carbon copy output of the training data.
forgive the skepticism, but this translates directly to "we asked the model pretty please not to do it in the system prompt"
Would it really be infeasible to take a sample and do a search over an indexed training set? Maybe a bloom filter can be adapted
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#146Earlier quoted context omitted.
There is still enormous value in cleaning up the long tail of somewhat important stuff. One of the great benefits of Claude Code to me is that smaller issues no longer rot in backlogs, but can be at least attempted immediately.
The difference is that Claude Code actually solves practical problems, but pure (as opposed to applied ) mathematics doesn't. Moreover, a lot of pure mathematics seems to be not just useless, but also without intrinsic epistemic value, unlike science. See https://news.ycombinator.com/item?id=46510353
Applications for pure mathematics can't necessarily be known until the underlying mathematics is solved.
Just because we can't imagine applications today doesn't mean there won't be applications in the future which depend on discoveries that are made today.
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#147Re: Erdos 281 solved with ChatGPT 5.2 Pro
#148Earlier quoted context omitted.
There is still enormous value in cleaning up the long tail of somewhat important stuff. One of the great benefits of Claude Code to me is that smaller issues no longer rot in backlogs, but can be at least attempted immediately.
The difference is that Claude Code actually solves practical problems, but pure (as opposed to applied ) mathematics doesn't. Moreover, a lot of pure mathematics seems to be not just useless, but also without intrinsic epistemic value, unlike science. See https://news.ycombinator.com/item?id=46510353
It's unclear to me what point you are making.
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#149Earlier quoted context omitted.
There is still enormous value in cleaning up the long tail of somewhat important stuff. One of the great benefits of Claude Code to me is that smaller issues no longer rot in backlogs, but can be at least attempted immediately.
The difference is that Claude Code actually solves practical problems, but pure (as opposed to applied ) mathematics doesn't. Moreover, a lot of pure mathematics seems to be not just useless, but also without intrinsic epistemic value, unlike science. See https://news.ycombinator.com/item?id=46510353
It's hard to know beforehand. Like with most foundational research.
My favorite example is number theory. Before cyptography came along it was pure math, an esoteric branch for just number nerds. defund Turns out, super applicable later on.
Re: Erdos 281 solved with ChatGPT 5.2 Pro
#150Earlier quoted context omitted.
forgive the skepticism, but this translates directly to "we asked the model pretty please not to do it in the system prompt"
It's mind boggling if you think about the fact they're essential "just" statistical models It really contextualizes the old wisdom of Pythagoras that everything can be represented as numbers / math is the ultimate truth
How so? Truth is naturally an apriori concept; you don't need a chatbot to reach this conclusion.