lonely boy trying to profit from the math ppls existential crisis
- Tweets 211
- Following 48
- Followers 21
- Likes 69
I don't care if you have 20 followers or 2000 If you're building, drop a comment I'm supporting you
16
TIL that my math PhD was a "mere manipulation of discrete symbols"
Pretty much the only domains in which current AI technology can invent solutions to new problems are those that:
1. Are easily expressed as sequences of discrete symbols
2. The mere manipulation of discrete symbol sequences can constitute
the substrate of reasoning.
3. The system can automatically check whether a generated solution is correct (without going through humans or real world experiments)
Today, this means theee domains: mathematics, coding, and situations can be accurately simulated with conventional software.
5
62
Can't reply because the brute blocked me, but look at he interacts with people on twitter who disagree with him.
1
53
Why did you also secretly make Astra worse? Comment below if you agree.
GPT-6 Sol and Luna are out. Not only are they a very significant improvement across the board, but also in writing and general "you know when you try it" quality.
We are also permanently reducing the API price by 50% making both of them viable for a ton of new usecases and making your usage go further too, even on the subscriptions.
And one more thing. We are loading a banked reset into all accounts of our Plus, Pro and Business users. Let's go!
openai.com/index/introducing…
2
57
Actually, it looks like Ricardo going manic. Does anyone here know him personally? Can someone do a wellness check?
2
86
i like to play math
math is not a sport - we are not playing games
It's like saying that an astronomer who uses a bigger telescope to see further is "doping" because he has access to better data
we need to get away from treating math like a sport where the only purpose is to show we are so clever
if we are just playing games, then the public funding agencies should defund us immediately and totally
1
31
it looks like all the arguments here are "I'm a professor with a fancy title at an ivy league institute, listen to me plz"
2
57
are any of my followers looking forward to this
🚨 GPT-6 Sol Drops Tomorrow
> GPT-6 Sol has an April 30 cutoff, same as Astra
> "Sol 5.6" is reportedly routing some selected users to Sol 6
> Early testers say Sol 6 is heavily RL-trained
> And apparently... it's insanely fast
> Luna could potentially launch alongside Sol
> Astra for the hardest tasks, Sol for everything else
Do you think it Can Beat Astra?
2
50
why did you have to mention the birth year?
Maggie Miller, born around 1993, still finds surprises in four-dimensional knots.
She earned a bachelor's at UT Austin and a 2020 Princeton PhD under David Gabai. Now assistant professor at UT, she studies knotted surfaces in 4-manifolds.
With collaborators she resolved a 1982 Livingston question by constructing Seifert surfaces that stay non-isotopic in the 4-ball. Her thesis advanced a 35-year Casson-Gordon problem on fibered ribbon knots.
Her honors include the 2023 Maryam Mirzakhani New Frontiers Prize, Forbes 30 Under 30, plus 2025 Sloan and Packard fellowships. She collects twisty puzzles and plays ukulele.
The simplest 4D questions, she shows, can still reshape topology.
34
how much effort has CV put into his latest math papers? i heard 0 lol . Prompting a co-author is no better than prompting an LLM
« Plus vous mettez de l’effort à surmonter le problème, plus vous êtes heureux » : Cédric Villani appelle les doctorants à ne pas utiliser l’IA générative
➡️ l.leparisien.fr/kzwt
27