Pedro Zuidberg Dos Martires
@pedrozudo.bsky.social
๐ค 459
๐ฅ 52
๐ 32
pedrozudo.github.io
Attending my forst tutorial of this year's UAI IN Rio. Fun fact about Rio. It's the only city outside of Europe that was the capital of a European countr at some point.
add a skeleton here at some point
2 months ago
0
0
0
Hot take: neural networks are the epicycles of the 21st century.
3 months ago
1
4
0
Ever changing from physics to AI and studying probabilistic ML, I have been wondering, how does all of this relate to quantum information theory ๐ฒ. I believe I've gotten a whole lot closer to the answer.
arxiv.org/abs/2506.01824
I'll be presenting the work at this year's
@auai.org
in Rio ๐
4 months ago
0
6
2
reposted by
Pedro Zuidberg Dos Martires
Jaron Maene
5 months ago
We developed a library to make logical reasoning embarrasingly parallel on the GPU. For those at ICLR ๐ธ๐ฌ: you can get the juicy details tomorrow (poster #414 at 15:00). Hope to see you there!
1
24
9
What is reasoning and have LLMs learned it? We answer these two questions in our new pre-print using the phase transition in random 3-SAT.
arxiv.org/abs/2504.039...
Together with
@rishihazra95.bsky.social
,
@gabventurato.bsky.social
and
@lucderaedt.bsky.social
Here is also a longer thread:
loading . . .
#ai #llms #reasoning #deepseek #3sat #iclr | Rishi Hazra
๐จ ๐ก๐ฒ๐ ๐ฃ๐ฎ๐ฝ๐ฒ๐ฟ ๐๐น๐ฒ๐ฟ๐: ๐๐ฎ๐๐ฒ ๐๐ฎ๐ฟ๐ด๐ฒ ๐๐ฎ๐ป๐ด๐๐ฎ๐ด๐ฒ ๐ ๐ผ๐ฑ๐ฒ๐น๐ ๐๐ฒ๐ฎ๐ฟ๐ป๐ฒ๐ฑ ๐๐ผ ๐ฅ๐ฒ๐ฎ๐๐ผ๐ป? Despite their impressive performance on reasoning tasks, LLMs often succeed by fitting to patterns rather than understanding logic. Are...
https://www.linkedin.com/posts/rishi-hazra-559b54224_ai-llms-reasoning-activity-7315306734682615808-Ep8D?utm_source=share&utm_medium=member_android&rcm=ACoAABrWbYgBamrBHHoxNy6HSFPMB7u480qCbmw
6 months ago
1
10
4
For years I have been studyibg the top researchers in AI. They all share one personal trait: attention to DETIAL.
6 months ago
0
1
0
I miss the times when all we cared about was end-to-end differentiability. Now all there seems to be is world domination this, world domination that.
7 months ago
0
2
0
Dieses Paradox schreit nach einem Namen.
add a skeleton here at some point
8 months ago
0
0
0
I think deepseek is amazing and I kind of like it... Buuuut... ๐ (Saw something like this in someone else's post but can't find it back)
8 months ago
0
1
0
Probability densities are like the sun. Look at them directly and they will blind you ๐
@paolomorettin.bsky.social
8 months ago
2
2
0
Is it still okay to justify gaps in my CV with the pandemic? Asking for a friend.
8 months ago
0
0
0
reposted by
Pedro Zuidberg Dos Martires
9 months ago
Are you interested in more scalable reasoning under uncertainty and attending NeurIPS? Then pass by our poster #3708 later today at 4.30pm! ๐ We use recursive integer arithmetic to express combinatorial problems and add uncertainty. Inference can be massively accelerated with tensors and the FFT. ๐
1
12
4
Recently I came across two papers that characterize modern machine learning as pseudoscience... They are probably right.
loading . . .
The reanimation of pseudoscience in machine learning and its ethical repercussions
Machine learning has a pseudoscience problem. In this perspective, the authors explore the recent resurgence of deep learning-assisted physiognomy and argue that pseudoscientific and socially harmful ...
https://www.cell.com/patterns/fulltext/S2666-3899(24)00160-0?_returnURL=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS2666389924001600%3Fshowall%3Dtrue&fbclid=IwY2xjawFaigtleHRuA2FlbQIxMAABHQNTjsLNaCbhUEyfNReAMYOemeZk6eZMnJBz627KUgEJKUET3SAtN2QH4g_aem_IhjU1jVkZm8s92CS4jdNlg
10 months ago
2
2
0
This place kind of feels sane๐ฆ๐
10 months ago
0
5
0
you reached the end!!
feeds!
log in