jrflo

5 ms·

1.
by jrflo 4d ago
1) there would be no incentive to develop a new model (non-distilled) if it had to be released as open weights. 2) frontier models are way more dangerous if they are open. It’s opening Pandora’s box, there’s no going back once they are rele
2.
by jrflo 5d ago
It's not only Lean code, there are English writeups too. To my understanding the pipeline for these problems is 1) solve in english 2) formalize systematically to check. No one is tackling problems purely in Lean, to my understanding.
3.
by jrflo 5d ago
That doesn't seem to be true. The OpenAI NS paper was 166 pages. Wiles-Taylor proof of Fermat's last theorem is 129 pages. The length is not unprecedented for a difficult unsolved problem. To be honest, I feel like the difficulty
4.
by jrflo 5d ago
I don't think it really attacks human understanding though. You can still read and understand an AI written proof. If another person comes up with a solution to a problem, you can read their methods and understand it. It doesn't m
5.
by jrflo 5d ago
I don't think that AI would disrupt any of that. Even if AI solves a problem, you can still discuss the methods at conferences, seminars, lectures, chats in the hallway, advising students, etc.
6.
by jrflo 5d ago
Right. It seems like reading an AI proof (although it may not be well written) will provide the same insights as reading a proof from another mathematician, assuming it's been reviewed and edited, just like any human-authored publicati
7.
by jrflo 6d ago
The circle jerking of Chinese models on this site never ceases to amuse me.
8.
by jrflo 6d ago
But who gets credit then? Every mathematician who's work was read by an LLM during training? By that logic, we should put every published mathematician's name on the authorship of this paper. Sure, this guy should be higher up the
9.
by jrflo 6d ago
I wasn't aware of that, definitely a shady practice if that's the case.
10.
by jrflo 6d ago
I pay for the Pro ChatGPT plan, and if you go to settings > data controls this is the first setting: > Improve the model for everyone > Allow your content to be used to train our models, which makes ChatGPT better for you and every
11.
by jrflo 7d ago
I could see their comment on user training data as a bit of a CYA statement, but removing Alpoge from the paper is awful. Has really soured what could have been a huge moment for AI progress.
12.
by jrflo 7d ago
The efficacy of applied NS was never in doubt. "Checking the box" is downplaying the magnitude of the discovery quite a bit as it has been unsolved for almost 100 years. Yes, this particular problem with NS no real-world applicati
13.
by jrflo 8d ago
I don't think he was defecting or leaking directly, just that it's entirely possible that this information got to OpenAI as a rumor rather than them directly spying on mathematicians chat logs.
14.
by jrflo 8d ago
I'm so tired of this "It's just marketing!!" commentary. An AI model just proved one of the top 3 unsolved problems in mathematics, they have a Lean certificate showing it's valid. How much more evidence do you need
15.
by jrflo 8d ago
I feel like it's far more likely that ordinary corporate espionage or leak led to this rather than OpenAI sifting through piles of user data to find this approach. Buckmaster's collaborator works at Anthropic, and could have been
16.
by jrflo 8d ago
To my understanding, those mathematicians proved a subset of problems, not the Navier-Stokes problem itself. OpenAI used that subproblem in its proof of NS it seems. The drama comes from where OpenAI got the idea to use that route to tackle
17.
by jrflo 8d ago
It's possible that OpenAI was mining prominent researcher's chats for inspiration to tackle these problems, but it's also entirely possible the leak came from his collaborator's end as he works at Anthropic and I'm
18.
by jrflo 8d ago
It's the tried and true headline method of "event 2 happened after event 1", implying a causal effect between the two, despite no evidence of such effect existing. Technically the headline is correct, in the same way that &qu
19.
by jrflo 11d ago
That's really cool, I would definitely be interested in a "normal" keyboard that had this kind of layout as a guitar player. Needing to remember different hand shapes to play a major chords took a long time to get used to - w
20.
by jrflo 12d ago
FYI you're trying to repair surface-mounted components by hand, that's pretty tricky. Most hand soldering is done with through hole components, not surface mounted. Not to say it can't be done, but you're starting on har
21.
by jrflo 12d ago
Holy shit, this has to be one of the most difficult proofs to formalize due to it's length and complexity right?
22.
by jrflo 12d ago
I think people are pissy because they're scared of AI and have consumed a lot of misinformation. Most people I know still think that datacenters are going to literally drain the lakes. No one is actually informed on the details, it&#x2
23.
by jrflo 12d ago
They already do, indirectly. The local government gets property taxes from these data centers. But maybe it would be more real for people if you just wrote them a check I guess.
24.
by jrflo 12d ago
So it's just a vscode wrapper around a mystery open weights LLM I'm guessing? > Can I choose which model Bob uses? > No. Bob automatically selects the most appropriate language model for each task based on complexity, requir
25.
by jrflo 13d ago
Idk, they’re trying to sell a $500/mo/seat service to tell you what model is best. I think it’s in their interest to keep it confusing and opaque. Not exactly independent.
26.
by jrflo 13d ago
Those robots are a gimmick and they can only do prescribed tasks in a super constrained environment. They are cashing in on LLM hype right now, vision has had some advances thanks to transformers but we are so far away in terms of the hard
27.
by jrflo 13d ago
You are describing superintelligence (ASI) not general intelligence (AGI)
28.
by jrflo 13d ago
You should read more on the ARC prize, it actually has a pretty long history. We're on the 3rd iteration because they keep getting saturated. If you look at the score history over time on ARC AGI 1, 2 and 3 it's pretty impressive.
29.
by jrflo 13d ago
So if you use the non-promotional price for sol it's only 25% higher?
30.
by jrflo 13d ago
Yeah I wonder what's going on, even when Anthropic soft launched Fable/Mythos I'm pretty sure they had model cards. Weird for GPT-6 to launch without a tweet from Altman too. I'm sure that one of the articles published p

More ›