Toutes les curations

6 curations · 1314 membres · fenêtre de 6 h, demi-vie 6 h, seuil 2 relayeurs

0liens
8tweets
88806posts lus
953auteurs
9206liens en base
1471 hde recul

Les tweets 8

Axe indépendant : repris sans regarder les liens.

1
Ramez Naam @ramez · 04:11
Anthropic is gaining exponentially less capability per billion dollars spent over time. It's now 8x as expensive to boost AI capabilities as it was 2 years ago, getting worse on a log scale. This includes METR report & internal models. Credit to @alexolegimas, @sebkrier, and
2 relayeurs à 2 · il y a 10 min Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
Gary Marcus @GaryMarcus a cité · 03:36
the end is near
2
R
Ryo Lu @ryolu_ · 02:14
when we can do anything, do we still know what we want? the full talk from @aiDotEngineer shanghai. every slide handmade in keynote, subs in 中文 and english
2 relayeurs à 2 · il y a 2 h Voir sur X
3
J
Justin Lo @loresjpg · 01:37
Nice DeepSWE score! Does it survive a lowercase filename? Introducing MaxxBench: 89 DeepSWE tasks with small changes to the instructions or starting code. Same engineering problem, slightly different requirement. Models clustered around 74% on DeepSWE end up 23 pp apart. Good
2 relayeurs à 14 · il y a 3 h Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
J
Julian Schrittwieser @Mononofu a cité · 01:37
Making novel variations of evals is very useful, public evals tend to get memorized!
Hamel Husain @HamelHusain a cité · 01:16
I see some are drawing the wrong inference from this study: they are focusing on this meta benchmark as yet another benchmark for model picking, i.e. open vs closed models, model families etc. That’s too narrow of an interpretation! You need to do your own private evals on
4
Greg Brockman @gdb · 01:55
frontier models, with lots of test time compute, are becoming the engine of scientific discovery
2 relayeurs à 2 · il y a 43 min Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
Gary Marcus @GaryMarcus a cité · 01:55
science ≠ math
5
S
Shaun Maguire @shaunmmaguire · 01:17
1) Jensen is one of the greatest entrepreneurs of this generation 2) he did something incredibly dumb 3) it was bad for America 4) but my guess is there was no bad intention 5) two mistakes don’t make a right, and I believe Jensen is irreplaceable for America’s AI leadership
2 relayeurs à 2 · il y a 3 h Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
Bing Xu @bingxu_ a cité · 01:17
+1 Jensen is the irreplaceable leader of 🇺🇸 AI. I'm proud of having worked directly or indirectly with him for the past 15 years. He's a true patriot who made a dumb mistake every man may make. We should continue to trust him.
6
J
Jameson Hodge @jamesonhodge · 01:01
I only make serious software
2 relayeurs à 2 · il y a 3 h Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
A
Alex Volkov @altryne a cité · 01:01
Be very very careful, if you watch this video your whole timeline is going to be this thing doing different songs.
Mat Velloso @matvelloso a cité · 08/11 22:41
This might be the best piece of software ever built
7
B
Bill Ackman @BillAckman · 01:06
An alert: @JensenHuang is human. And he is an incredible human. Even incredible humans make mistakes. He takes full responsibility as he should. We should all support him and his family. Onward.
2 relayeurs à 2 · il y a 3 h Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
Rohan Paul @rohanpaul_ai a cité · 08/11 22:30
🎯🎯
8
T
tae kim @firstadopter · 00:11
Nvidia CEO Jensen Huang's email to Nvidia employees: To my colleagues at NVIDIA, I owe my family and all of you an apology. I was unfaithful to my wife. My actions hurt her and our children. There is no excuse for what I did. I am deeply sorry. Throughout my life, my family
2 relayeurs à 2 · il y a 4 h Voir sur X
Qui l'a repris
Ce qu'ils en ont dit
Sam Bhagwat @calcsam a cité · 00:11
This reminds me of a story a mentor of mine told me 15 years ago, that his dad (Robert) told him. Robert grew up in northwest Washington. He and his high school sweetheart both had horrible family situations. They decided to elope the day after their graduation. This was