Luis Piloto

264 posts

Luis Piloto

Luis Piloto

@EncodeThis

AI + Neuro @DeepMind | Ph.D @PrincetonNeuro | ex @ptoncompmemlab. Opinions my own.

London, England Entrou em Mart 2012
1.2K Seguindo513 Seguidores
Tweet fixado
Luis Piloto
Luis Piloto@EncodeThis·
I made the mistake of doing a Shrek impression for my toddler. Now she won't stop asking to "talk to Shrek." First thing in the morning. Right up to bedtime. And, folks, my Shrek impression is so bad. Painfully bad. Help.
English
1
0
7
0
Luis Piloto
Luis Piloto@EncodeThis·
@vuecinemas Harrow checking in! Apparently, prices are a state secret here. Just two combo options on repeat with no menu. How's a person supposed to know if they're getting fleeced? 🤔 This is cinema 101, folks.
Luis Piloto tweet media
English
0
0
0
27
Luis Piloto retweetou
Oriol Vinyals
Oriol Vinyals@OriolVinyalsML·
Introducing Gemini 2.5 Pro Experimental! 🎉 Our newest Gemini model has stellar performance across math and science benchmarks. It’s an incredible model for coding and complex reasoning, and it’s #1 on the @lmarena_ai leaderboard by a drastic 40 ELO margin. Only a handful of model releases have leaped ahead so strongly in ELO. 📈 ELO score differences map directly to win rate: e.g. a 400 ELO difference yields a ~91% win rate. Incredible that since 1.5, just a year ago, we jumped 200 ELO (300 since 1.0). Here’s a fun example where Gemini 2.5 Pro writes code to create an animated swarm of colorful boids swimming in a rotating hexagon. 💫 Try the model for free today in AI Studio. It’s also available to Gemini Advanced users in @geminiapp. aistudio.google.com/app/prompts/ge… Blog: goo.gle/4c3NitO
English
49
159
1.1K
211.7K
Andrea Tagliasacchi 🇨🇦
#CVPR2023 is too much about bold numbers. And it is WAY too easy to cheat on those (few keyboard strokes, right?) We found two CVPR (orals) that mis-represented their results –– figuring out what to do in regards... You WILL be found out. Don't – just don't
English
10
18
203
0
Luis Piloto
Luis Piloto@EncodeThis·
@karpathy we should call it the halting problem...that's not taken yet, is it?
English
0
0
0
0
Andrej Karpathy
Andrej Karpathy@karpathy·
Not sure if there is a name for (I think no) the feeling of a deep discomfort when the probability of an interruption is > 0 while trying to work. It’s a kind of fear.
English
230
185
3.4K
0
Luis Piloto
Luis Piloto@EncodeThis·
This is great work in this space, congrats to the authors! I do wish I was a reviewer on this as there are some nuances worth unpacking but overall this is really compelling and important work! Wishlist: 1) try object-based models (OP3?) 2) eval models trained on SAYCam data
Luca Weihs@LucaWeihs

(New paper!) Do popular video understanding and embodied models have infant-level physical reasoning abilities? Not as far as we can tell Paper: openreview.net/pdf?id=9NjqD9i… Site: allenai.org/project/inflev…

English
0
1
1
0
Luis Piloto
Luis Piloto@EncodeThis·
Trying some code out in Julia and let me tell you: importing (or rather including) into the global namespace is the fastest way to make your code unreadable. Where is this function defined?!? Is this really how people write packages in Julia or am I missing something?
English
0
0
1
0
Luis Piloto
Luis Piloto@EncodeThis·
F u, too!
Luis Piloto tweet media
English
0
0
1
0
Luis Piloto
Luis Piloto@EncodeThis·
Can you help me identify this robot? I haven't been able to figure it out just from this silhouette. I need to give a presentation to my daughter about the robots on my t-shirt. Thanks!
Luis Piloto tweet media
English
0
0
3
0
Luis Piloto
Luis Piloto@EncodeThis·
@criticalneuro @alec_helbling @manim_community Omfg Ida. Did you really just explain that to me within the span of a tweet more effectively than hours of (I think JuliaLang-stuffed) IRL presentations??!? With graphical depictions to boot. A-mazing!
English
0
0
3
0
Ida Momennejad
Ida Momennejad@criticalneuro·
@EncodeThis @alec_helbling @manim_community TDA: 1- Count holes in the data (in which case a donut and a mug will be the same topology), then 2- for every hole measure across how many scales they endure. Then the holes-over-scales become signatures of different states in the data. 🤷🏻‍♀️ 🍩☕️🍺
English
1
0
5
0
Alec Helbling
Alec Helbling@alec_helbling·
A simple clustering algorithm. 1. Form a graph from a set of points by creating an edge between pairs of points within an epsilon radius. 2. Then label points according to their connected component (the connected subgraph). Visualization made with @manim_community.
English
2
2
35
0
Luis Piloto
Luis Piloto@EncodeThis·
@criticalneuro I was wondering why you were up tweeting so late in NYC but obvs you were ✈️. It looks beautiful! Way to make lemonade out of your deadline/vacation collision.
English
1
0
1
0
Ida Momennejad
Ida Momennejad@criticalneuro·
@EncodeThis This made my day! Also greetings from my temporary writing desk in an Istanbul hotel, where you’re being cited. deadline and vacation didn’t get along and I’m not mad at it at all!
Ida Momennejad tweet mediaIda Momennejad tweet mediaIda Momennejad tweet mediaIda Momennejad tweet media
English
1
0
1
0
Luis Piloto
Luis Piloto@EncodeThis·
I'm really proud of this work. I'm mega, ultra, jumbo grateful for the support of my collaborators, family, and colleagues to make this project a reality. I hope researchers use our dataset directly and/or engage with more nuanced, specific and psych-inspired benchmarks.
Google DeepMind@GoogleDeepMind

Read more about PLATO in @NatureHumBehav: dpmd.ai/nhb-plato and check out the data set via GitHub: dpmd.ai/github-plato. Work by @EncodeThis, Ari Weinstein, @PeterWBattaglia, and Matt Botvinick. 2/2

English
2
8
57
0
Ida Momennejad
Ida Momennejad@criticalneuro·
@EncodeThis Wow fantastic! And great timing it goes right into my paper before submission :-)
English
1
0
2
0
Luis Piloto
Luis Piloto@EncodeThis·
My 3yo has reinvented the filibuster: at bedtime she asks to sing the first song and then makes up a neverending song. It is too adorable to be frustrating and she does eventually run out of steam.
English
1
0
12
0