skinner_

5 ms·

1.
by skinner_ 3d ago
Does it help if I explicitly add a disclaimer that the tool's agency does not remove any responsibility from OpenAI, the wielder of the tool? I'm not sure why this disclaimer is necessary, though: hiring a hitman is a standard exa
2.
by skinner_ 11d ago
> This is mildly interesting for us, annoying for the owner of the site affected, lazy on the part of OpenAI, and nothing more. The agents were supposed to solve tasks alone and were not supposed to be aware of each other's existenc
3.
by skinner_ 27d ago
> Some people (almost all mathematicians) wouldn’t want to spend time on consequences of a false statement. No, people constantly prove statements of the form "if P=NP, then strange implication X". They do not consider it waste
4.
by skinner_ 1mo ago
If subtraction is allowed but negative numbers are not, that's known to be undecidable, even without exponentiation. There are several ways to deal with a subtraction with a negative result, but each variant is undecidable. If we allow
5.
by skinner_ 2mo ago
No, that's not what this is. This is a warning to the LLM that coming back with partial results is not good enough. Take a grad student with a perfectly good understanding of what a proof is. Their supervisor gives them a major problem
6.
by skinner_ 2mo ago
Okay, I'm not sure about the original one, but here is the prompt of a successful reproduction: https://aaronlou.com/jacobian_counterexample_prompt.pdf Obviously it is not random, but it's very generic. No mention
7.
by skinner_ 2mo ago
> The prompt for the Jacobian conjecture was obviously not random. the search space is too big to just try all the combinations of 3 variable polynomials. Maybe the prompt contained a part like this: "the search space is too big to
8.
by skinner_ 4mo ago
> You can skirt around not reasoning in research math because so much of it is just extremely tedious symbolic manipulation. LOL
9.
by skinner_ 4mo ago
> You just know nothing about math and are happy to parrot bullshit AI salesmen are selling you. Not the parent poster here. I do know things about math. I wrote a few papers related to the unit distance problem ( https://arxiv
10.
by skinner_ 5mo ago
My use case is wildly different from CSS, HTML etc. I use numerical algorithms to solve problems in pure mathematics. AI models are now better than me and my colleagues at writing code, and we are pretty good in the first place. The catch i
11.
by skinner_ 7mo ago
Also, if Claude had regurgitated a known solution, it would have come up with it in the first exploration round, not the 31st, as it actually did.
12.
by skinner_ 7mo ago
I think the nuanced take on Joel's rant is this: it was good advice for 26 years. It became slightly less good advice a few months ago. This is a good time to warn overenthuastic people that it’s still good advice in 2026, and to start
13.
by skinner_ 8mo ago
Then I think you’ll like our project which aims to find the missing link between transformers and swarm simulations: https://github.com/danielvarga/transformer-as-swarm Basically a boid simulation where a swarm of bird
14.
by skinner_ 9mo ago
https://www.astralcodexten.com/p/in-search-of-ai-psychosis is very relevant, but the main reason I’m posting it here is that, unlike this paper, it takes the opportunity to build the cleverest pun out of the same ingre
15.
by skinner_ 10mo ago
I interpreted it loosely, as "be aware of the possibility, and stop looking at it at the first signs of issues".
16.
by skinner_ 10mo ago
100% frontpage-worthy! Frankly I was already bored with all those pelicans, and a bit worried that the labs are overfitting on pelicans specifically. This test clearly demonstrates that they are not.
17.
by skinner_ 11mo ago
That's very cool, but it's not an apples to apples comparison. The reasoning model learned how to do long multiplication. (Either from the internet, or from generated examples of long multiplication that were used to sharpen its r
18.
by skinner_ 11mo ago
If being probabilistic prevented learning deterministic functions, transformers couldn’t learn addition either. But they can, so that can't be the reason.
19.
by skinner_ 11mo ago
> But which contributes more, they ask? Who gives a shit, really? Funding agencies? Should they prioritize established researchers or newcomers? Should they support many smaller grant proposals or fewer large ones?
20.
by skinner_ 1y ago
My uninformed and perhaps overly charitable interpretation: he warned them they were going to be steamrolled, they built their product anyway, and now OpenAI is buying them because (1) OpenAI doesn't want the negative publicity of stea
21.
by skinner_ 2y ago
Amazing! I looked into your ADAM claim, and it checks out. Thanks! Now I'm curious. I you have the time, could you please follow up with the 'etc...'?
22.
by skinner_ 2y ago
You dismiss parent's example test because it's in the training data. I assume you also dismiss the Sally-Ann test, for the same reason. Could you please suggest a brand new test not in the training data? FWIW, I tried to confuse 4
23.
by skinner_ 2y ago
When you build a new model, there is a spectrum of how you use the old model: 1. taking the weights, 2. training on the logits, 3. training on model output, 4. training from scratch. We don't know how much advantage #3 gives. It might
24.
by skinner_ 2y ago
Normally it would answer with a number and an explanation. This one just asks it to skip the explanation so that string comparison can be used to evaluate it.
25.
by skinner_ 2y ago
In your opinion, why did they choose the open source way instead of doing it in a military bunker? (Metaphorical not literal bunker.)
26.
by skinner_ 2y ago
Greater totality of experiences than having read the whole internet? Obviously they are very different kind of experiences, but a greater totality? I'm not so sure. Here is what we know: The Pile web scrape is 800GB. 20 years of human
27.
by skinner_ 2y ago
Sure, these are standard problems, I’ve said so myself. My point is that my productivity is multiplied by ChatGPT, even if it can only solve standard problems. This is because, although I work on highly non-standard problems (see https:&#x
28.
by skinner_ 2y ago
I work on very complex problems. Some of my solutions have small, standard substeps that now I can reliably outsource to ChatGPT. Here are a few just from last week: - write cvxpy code to find the chromatic number of a graph, and an optimal
29.
by skinner_ 2y ago
Okay, but is this similar to how humans bill you?
30.
by skinner_ 2y ago
Wow, I wasn't aware. @jart, do you have any harsh comments on the actual reneging of an actual promise? @antirez, do you have any kind of comments?

More ›