Alexia Jolicoeur-Martineau @jm_alexia
AI Researcher at the Samsung SAIT AI Lab 🐱💻 ajolicoeur.wordpress.com Montréal, Québec Joined March 2017-
Tweets7K
-
Followers10K
-
Following1K
-
Likes29K
📢📢Most diffusion (and flow matching) models use handcrafted schedules for their denoising steps during sampling. We show how to optimize them in a principled manner for high-quality generation! @amsabour added quickstart guide & collab to get you started quickly (links below)!
📢📢Most diffusion (and flow matching) models use handcrafted schedules for their denoising steps during sampling. We show how to optimize them in a principled manner for high-quality generation! @amsabour added quickstart guide & collab to get you started quickly (links below)!
My favorite part is that it works really well with out-of-the-distribution garments
My favorite part is that it works really well with out-of-the-distribution garments https://t.co/UD9frTnxyl
last thursday, Meta dropped Llama 3, the OpenAI killer. no doubt a very impressive model! but over the weekend, we discovered an extremely trivial programmatic jailbreak against llama 3...sorry zuck!😘 so much for all that safety-tuning☹️ code: github.com/haizelabs/llam…
🆕 Introducing JAT, the first open-source multi-modal, multi-task multi-domain agent! 🤖 A step toward open generalist agents! 🚀 📰 Blog: huggingface.co/blog/jat
Llama-3-8b already dethroned? The benchmarks look really good! Their 7b model is apparently significantly better Llama-3-8b!! 👀👀 Really excited to try this model out, hope it gets released soon!
Llama-3-8b already dethroned? The benchmarks look really good! Their 7b model is apparently significantly better Llama-3-8b!! 👀👀 Really excited to try this model out, hope it gets released soon! https://t.co/G5PDc5NGlA
5T tokens FineWeb dataset just dropped @huggingface It's a 275GB dataset with cleaned and deduplicated data under an Open Data Commons license. We all see the difference the 15T tokens pre-training made for LLaMA-3 and now everyone can have it .
Zuck releasing a billion dollar model is actually wild, like really undermining what OAI is doing. flexing compute like “yea we can do that not a big deal”
Yes, we observed the same in our experiments on instruction-tuning. We need 8 times more "difficult" data for 1.3B model than 13B model to achieve a similar performance trend. Another key point we noted is, we only need difficult data. Easy data can just be ignored. I believe a…
Yes, we observed the same in our experiments on instruction-tuning. We need 8 times more "difficult" data for 1.3B model than 13B model to achieve a similar performance trend. Another key point we noted is, we only need difficult data. Easy data can just be ignored. I believe a… https://t.co/419oRhtvMb
Llama3 reminds everyone of the misconception about scaling laws again: it's not that a larger model is always better, but that a larger model is cheaper to train if you want to reach the same performance. Yes, this might be somewhat counter-intuitive, but this is one of the key…
Llama3 was trained on 15 trillion tokens of public data. But where can you find such datasets and recipes?? Here comes the first release of 🍷Fineweb. A high quality large scale filtered web dataset out-performing all current datasets of its scale. We trained 200+ ablation…
Llama3 was trained on 15 trillion tokens of public data. But where can you find such datasets and recipes?? Here comes the first release of 🍷Fineweb. A high quality large scale filtered web dataset out-performing all current datasets of its scale. We trained 200+ ablation…
ScaleFold: Reducing AlphaFold Initial Training Time to 10 Hours arxiv.org/abs/2404.11068
This incredibly well-written paper illustrates how deep domain thinking can lead to simplicity and why representation still matters with today’s AI. Also, this is how one should approach writing surveys. Not “Foo did X and Bar did Y. End of story.” arxiv.org/abs/2404.11735
Do you guys realize how wild this is? When it's done training it's gonna be one of the best LLMs, period. Better than most proprietary models too, probably even GPT-4!! Open models FTW y'all
Do you guys realize how wild this is? When it's done training it's gonna be one of the best LLMs, period. Better than most proprietary models too, probably even GPT-4!! Open models FTW y'all
The conclusive EagleX is here Based on the RWKV-v5 architecture, bringing into opensource 7B space, the best SOTA - Multi-lingual model - English perplexity model - Attention-free transformer today (10-100x+ lower inference) With comparable English performance to Mistral
you're telling me an 8B param model was trained on fifteen trillion tokens? i didn't even know there was that much text in the world really interesting to see how scaling laws have changed best practices; GPT-3 was 175 billion params and trained on a paltry 300 billion tokens
🥁 Llama3 is out 🥁 8B and 70B models available today. 8k context length. Trained with 15 trillion tokens on a custom-built 24k GPU cluster. Great performance on various benchmarks, with Llam3-8B doing better than Llama2-70B in some cases. More versions are coming over the next…
Je vais à Montréal! This June I'm starting a new position as an assistant professor at @UMontreal and as a core academic member of @Mila_Quebec. Drop me a line if you're interested in working together on problems in AI4Science, Optimal Transport, and Generative Modeling.
After two good years at Microsoft Research AI4Science, I am very excited to announce that as of this month I have, together with Chad Edwards, co-founded a new startup in the field of molecular and materials discovery.
Anyways, I do find the results of this paper (authored with @VikaBsmv and @rtsarfaty) to be interesting and worthwhile. Here is a summary of the experimental conditions and results. Models fail on very simple inferences, and more so in embedded sentential contexts.
AK @_akhaliq
309K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo follow on Hugging Face: https://t.co/q2Qoey80Gx
Peyman Milanfar @docmilanfar
67K Followers 260 Following Distinguished Scientist at Google Research. Computational Imaging, Machine Learning, and Vision. Tweets = personal opinions. May change or disappear over time.
Delip Rao e/σ @deliprao
46K Followers 5K Following Busy inventing the shipwreck. @Penn. Past: @johnshopkins, @UCSC, @Amazon, @Twitter ||Art: #NLProc, Vision, Speech, #DeepLearning || Life: 道元, improv, running 🌈
Kevin Patrick Murphy @sirbayes
42K Followers 334 Following Research Scientist at Google Brain / Deepmind. Interested in Bayesian Machine Learning.
Rosanne Liu @savvyRL
33K Followers 966 Following Cofounded & running @ml_collective. Host of Deep Learning Classics & Trends. Research at Google DeepMind. DEI/DIA Chair of ICLR & NeurIPS. Writing https://t.co/IbycyGfnDR
Horace He @cHHillee
23K Followers 448 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemale
Sander Dieleman @sedielem
50K Followers 2K Following Research Scientist at Google DeepMind. I tweet about deep learning (research + software), music, generative models (personal account).
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
Dmytro Mishkin 🇺�.. @ducha_aiki
18K Followers 591 Following Marrying classical CV and Deep Learning. I do things, which work, rather than being novel, but not working.
Tanishq Mathew Abraha.. @iScienceLuvr
54K Followers 1K Following PhD at 19 | Founder and CEO at @MedARC_AI | Research Director at @StabilityAI | @kaggle Notebooks GM | Biomed. engineer @ 14 | TEDx talk➡https://t.co/xPxwKTq6Qb
Irina Rish @irinarish
9K Followers 995 Following prof UdeM/Mila; Canada Excellence Research Chair; AAI Lab head https://t.co/UzlrC7ZrGF; INCITE project PI https://t.co/0rV7szd7rH; CSO https://t.co/XDhj6MEtUj
Kyunghyun Cho @kchonyc
61K Followers 2K Following a combination of a mediocre scientist, a mediocre manager, a mediocre advisor & a mediocre PC at @nyuniversity (@CILVRatNYU) & @genentech (@PrescientDesign).
Tom Goldstein @tomgoldsteincs
23K Followers 2K Following Professor at UMD. AI security & privacy, algorithmic bias, foundations of ML. Follow me for commentary on state-of-the-art AI.
Michael Bronstein @mmbronstein
43K Followers 4K Following #DeepMind Professor of #AI @UniofOxford / Fellow @ExeterCollegeOx / ML Lead @ProjectCETI / https://t.co/kZpGpDzYeV
rohan anil @_arohan_
12K Followers 2K Following Principal Engineer, @GoogleDeepMind Gemini. prev PaLM-2. Tinkering with optimization and distributed systems. opinions are my own.
Sara Hooker @sarahookr
39K Followers 7K Following I lead @CohereForAI. Formerly Research @Google Brain @GoogleDeepmind. ML Efficiency at scale, LLMs, @trustworthy_ml. Changing spaces where breakthroughs happen.
Ferenc Huszár @fhuszar
40K Followers 1K Following Secular Bayesian. Associate Professor in Machine Learning @Cambridge_CL. Talent aficionado at https://t.co/RbJkoLguey Alum of @Twitter, Magic Pony and @Balderton
Tim Dettmers @Tim_Dettmers
29K Followers 819 Following PhD Student at @UW. I blog about deep learning and PhD life at https://t.co/Y78KDJJFE7.
Miles Brundage @Miles_Brundage
43K Followers 10K Following Policy research at @openai. I mostly tweet about AI, animals, and sci-fi. He/him. Views my own.
Sophia Bliss @SophiaBlis36884
11 Followers 834 Following
강은희 @ehk_kor
0 Followers 4 Following
AI & Semi Fab News @SemiFabAI
4 Followers 246 Following This is my account for tracking financial/dev news in the Artificial-Intelligence and high-tier semi-conductor fabrication sectors.
Pankaj Gupta @pankaj_ipynb
26 Followers 919 Following The English language can not fully capture the depth and complexity of my thoughts. So I'm incorporating Emoji into my speech to better express myself 😉.
Jonathan Wang @givemettt5600
15 Followers 179 Following
DaisyGregory @Daisy243564281
63 Followers 4K Following interpol is connecting over 195 country for a safer word.Don't Hesitate To Report Serious Cyber Abuse or Suspicious Activities
Ronald Martino (Marti.. @ronald_mar87278
60 Followers 544 Following
Teng Xiao @TengX6
54 Followers 484 Following PhD student at Penn State University. Research Interest: Machine Learning
Melon @Ma_praew
578 Followers 2K Following Success is not final, failure is not fatal: It is the courage to continue that counts.
Nicolas @nicopipme
13 Followers 232 Following
Jeevak_Shetty @jeevak_she55933
4 Followers 1K Following
Nova @Nova274393
1 Followers 187 Following I swear in the name of God, don't miss an opportunity to earn 500-5000usdc every day. https://t.co/ZOnZK8jjjs
Measurer Star @feigaobox
142 Followers 1K Following
reddy_15 @15_reddy68201
4 Followers 958 Following
Luis @Luis81263596
50 Followers 94 Following
Stephan Mandt @StephanMandt
2K Followers 555 Following ML Professor @UCIrvine, previously @blei_lab, @Princeton. #GenerativeAI, #Compression, #AI4Science. Program Chair @aistats_conf 2024; General Chair AISTATS 2025
Mehdi Inane @MehdiInane
15 Followers 40 Following
Aaditya ; @Aaditya26082004
519 Followers 7K Following CS'26 • Machine Learning • Open-Source • Web Dev. • Algorithms • Jai Shree Krishna 🦚🪈
AI Daily Guy @Jakeharr
887 Followers 2K Following Founder of AffordHunt, the platform connecting indie hackers & SMBs with quality, budget-friendly alternatives. 💯 Follow Back #indiehacker #buildinpublic
Chen F.C. @fuchen2624
287 Followers 2K Following Indie Hacker, building SaaS products. AI Enthusiasm. follow and DM to get my support
Andrew white @Andreww95636515
130 Followers 3K Following 3d modeling. Gaussian splatting, NeRF, Diffusion models, GANs.
Kevin Sison @kevinsison
47 Followers 309 Following
Elon musk investment @tasla_investmen
67 Followers 634 Following cryptocin, Bitcoin, Tesla investment, xal investment.
Danilo J. Rezende @DaniloJRezende
35K Followers 1K Following Director @ #DeepMind Building models to accelerate fundamental sciences Prev @EPFL IFT @Polytechnique @fisicaUSP
Satyam Soni @SatyamSoni44642
5 Followers 60 Following
Zinnia @Zinnia821305
16 Followers 1K Following
Science With Sanjay @SanjayScience
113 Followers 1K Following Analytical Chemistry PhD Student | Purdue University | Optimization, Machine Learning, LLMs, and Automation | Former High Stakes Poker Player |
Dongjun Kim @gimdong58085414
699 Followers 712 Following PostDoc at Stanford; Diffusion models; My own words
Rabeh Boudia @RabehBoudiaAI
48 Followers 393 Following 🌟 here for the memes & AI, software development
Tom Cruise @CruiseFededi
148 Followers 379 Following
rosebeats09 @rosebeats09
178 Followers 448 Following Pain&Dark Beats https://t.co/XjOq0h6ife [email protected]
andrea @aerdnasan
230 Followers 900 Following photographer. ig : aerdnasan https://t.co/qwYvc9siDN co-founder @intenddot
Miltos Kofinas @MiltosKofinas
543 Followers 301 Following PhD student @UvA_Amsterdam | Graph Neural Networks | Geometric Deep Learning | Neural Fields | Spatiotemporal Forecasting
Su @BilgeSuuuu
1K Followers 1K Following
Daniel Han @danielhanchen
7K Followers 929 Following Building @UnslothAI. Finetune LLMs 30x faster https://t.co/aRyAAgKOR7. Prev ML at NVIDIA. Hyperlearn used by NASA. I like maths, making code go fast
sonorch @sonorchus
64 Followers 283 Following ˗ˏˋ ★ ˎˊ˗ ˗ˏˋ ★ ˎˊ˗ ˗ˏˋ ★ ˎˊ˗ ˗ˏˋ ★ ˎˊ˗ ˗ˏˋ ★ ˎˊ˗ ˗ˏˋ ★ ˎˊ˗
AK @_akhaliq
309K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo follow on Hugging Face: https://t.co/q2Qoey80Gx
Peyman Milanfar @docmilanfar
67K Followers 260 Following Distinguished Scientist at Google Research. Computational Imaging, Machine Learning, and Vision. Tweets = personal opinions. May change or disappear over time.
Soumith Chintala @soumithchintala
185K Followers 877 Following Cofounded and lead @PyTorch at Meta. Also dabble in robotics at NYU. AI is delicious when it is accessible and open-source.
Lucas Beyer (bl16) @giffmana
56K Followers 445 Following Researcher (Google DeepMind/Brain in Zürich, ex-RWTH Aachen), Gamer, Hacker, Belgian. Mostly gave up trying mastodon as [email protected]
Alfredo Canziani @alfcnz
86K Followers 269 Following Musician, math lover, cook, dancer, 🏳️🌈, and an ass prof of Computer Science at New York University
Delip Rao e/σ @deliprao
46K Followers 5K Following Busy inventing the shipwreck. @Penn. Past: @johnshopkins, @UCSC, @Amazon, @Twitter ||Art: #NLProc, Vision, Speech, #DeepLearning || Life: 道元, improv, running 🌈
Kosta Derpanis @CSProfKGD
48K Followers 198 Following #CS Associate Prof @YorkUniversity, #ComputerVision Scientist Samsung #AI, @VectorInst Faculty Affiliate, TPAMI AE, #CVPR2024/#ECCV2024 Publicity Co-chair
Michael Black @Michael_J_Black
58K Followers 638 Following Director, Max Planck Institute for Intelligent Systems (@MPI_IS). Chief Scientist @meshcapade. Building 3D digital humans using vision, graphics, and learning.
(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K Following
Rosanne Liu @savvyRL
33K Followers 966 Following Cofounded & running @ml_collective. Host of Deep Learning Classics & Trends. Research at Google DeepMind. DEI/DIA Chair of ICLR & NeurIPS. Writing https://t.co/IbycyGfnDR
Horace He @cHHillee
23K Followers 448 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemale
Sander Dieleman @sedielem
50K Followers 2K Following Research Scientist at Google DeepMind. I tweet about deep learning (research + software), music, generative models (personal account).
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
Jürgen Schmidhuber @SchmidhuberAI
106K Followers 0 Following Invented principles of meta-learning (1987), GANs (1990), Transformers (1991), very deep learning (1991), etc. Our AI is used many billions of times every day.
Dmytro Mishkin 🇺�.. @ducha_aiki
18K Followers 591 Following Marrying classical CV and Deep Learning. I do things, which work, rather than being novel, but not working.
François Fleuret @francoisfleuret
31K Followers 455 Following Prof. @Unige_en, Adjunct Prof. @EPFL_en, Research Fellow @idiap_ch, co-founder @nc_shape. AI and machine learning since 1994. I like reality.
Tanishq Mathew Abraha.. @iScienceLuvr
54K Followers 1K Following PhD at 19 | Founder and CEO at @MedARC_AI | Research Director at @StabilityAI | @kaggle Notebooks GM | Biomed. engineer @ 14 | TEDx talk➡https://t.co/xPxwKTq6Qb
Jia-Bin Huang @jbhuang0604
51K Followers 285 Following Associate Professor @umdcs; Part-time Research Scientist @Meta. I like pixels.
Dan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)
Maksym Andriushchenko.. @maksym_andr
3K Followers 933 Following phd student at @EPFL🇨🇭 // google & open phil phd ai fellow // past @adoberesearch @uni_tue // best way to support 🇺🇦 https://t.co/fxomgJ7NU9
Zhengyao Jiang @zhengyaojiang
1K Followers 261 Following Cofounder and CTO @WecoAI, building AutoML Agents. Final year PhD student at UCL @UCL_DARK @ai_ucl. (Zheng=j-uhng, j as in job; yao=y-aoww)
Robin Rombach @robrombach
6K Followers 397 Following Generative enthusiast and long-term PhD Student @LMU_Muenchen. Author of VQGAN, Latent Diffusion, Stable Diffusion.
RWKV @RWKV_AI
2K Followers 3 Following AI model built by the community, for everyone in this world Part of the Linux Foundation, Apache 2 licensed An RNN scaled to 14B params with GPT-level of perf
Alex Clemmer 🔥🔥.. @hausdorff_space
4K Followers 1K Following Brexit, Britney Spears, and Buffalo Wild Wings. if there aint no ring on my finger you aint goin on my gram wasq'u descendent.
African Institute for.. @AIMS_Next
20K Followers 1K Following AIMS is a pan-African network of Centres of Excellence for postgraduate training, research & public engagement in mathematical sciences & STEM.
AIMS RWANDA @AIMS_Rwanda
4K Followers 202 Following 5th Centre @AIMS_Next, a pan-African network of Centres of Excellence for postgraduate training, research & public engagement in mathematical sciences & STEM.
Taelin @VictorTaelin
17K Followers 900 Following Founder of @HigherOrderComp Building the massively parallel future of computing Reaching AGI to cure all diseases and suffering is all that matters
Balatro @BalatroGame
17K Followers 18 Following Balatro is a poker-inspired roguelike deckbuilder where you play poker hands and earn chips to defeat enemy blinds. OUT NOW on Steam, PS, Switch & Xbox.
localthunk @LocalThunk
20K Followers 43 Following Solo Developer and Artist for @BalatroGame Business and Media Inquiries: [email protected]
Aaron Defazio @aaron_defazio
6K Followers 359 Following Research Scientist at Meta working on optimization. Fundamental AI Research (FAIR) team
Dongjun Kim @gimdong58085414
699 Followers 712 Following PostDoc at Stanford; Diffusion models; My own words
Moin Nadeem @moinnadeem
2K Followers 979 Following Co-Founder at Phonic. Previously @Stanford CS PhD Dropout, @MosaicML, CS @MIT. I tend to be wrong, but the learning process makes it enjoyable. 🇵🇰🇺🇲
Noland Arbaugh @ModdedQuad
86K Followers 13 Following P1. Cyborg. Neuralnaut. Skynet Progenitor. Taurus.
Miltos Kofinas @MiltosKofinas
543 Followers 301 Following PhD student @UvA_Amsterdam | Graph Neural Networks | Geometric Deep Learning | Neural Fields | Spatiotemporal Forecasting
pleias @pleiasfr
207 Followers 1 Following
Jirka⚡Borovec @JirkaBorovec
730 Followers 536 Following Machine learning and Data science researcher focusing on computer vision...
Ivan Yashchuk @IvanYashchuk
366 Followers 738 Following I work on @PyTorch at @NVIDIA Born in Ukraine | Raised in Siberia Yugra | Living in Finland 🇫🇮
Luca Antiga ⚡️ @lantiga
3K Followers 2K Following CTO @LightningAI // Co-founder @ Orobix · Tensorwerk // Manning author
MedARC @MedARC_AI
4K Followers 10 Following Medical AI Research Center (MedARC) Unlocking new possibilities in medical AI research. Founded by @iScienceLuvr
Daniel Han @danielhanchen
7K Followers 929 Following Building @UnslothAI. Finetune LLMs 30x faster https://t.co/aRyAAgKOR7. Prev ML at NVIDIA. Hyperlearn used by NASA. I like maths, making code go fast
Hannes Stärk @HannesStaerk
8K Followers 331 Following @MIT PhD student • ML for molecular biology and flow generative models
Gabriele Corso @GabriCorso
4K Followers 637 Following PhD student @MIT • Research on Generative Models and Geometric Deep Learning for Biophysics • BA @CambridgeUni • Former @TwitterResearch, @DEShawGroup and @IBM
cloud @cloud11665
5K Followers 1K Following SIMD fan | ctf player | ex OI-er | accelerate. CEO and co-founder @figura_labs DM FOR PRIVATE BETA ACCESS
Wu Lin @LinYorker
169 Followers 13 Following Postdoctoral fellow at @VectorInst. ML PhD at UBC. Mathematical and computational structures for ML. Geometric and algebraic methods.
Emtiyaz Khan @EmtiyazKhan
11K Followers 234 Following Team leader at @RIKEN_AIP_EN. Opinions my own. Follow me at https://t.co/jXDOS1HKXE
Yufeng Zheng @YufengZzzz
245 Followers 49 Following
Xu Chen @XuChen71058062
778 Followers 303 Following PhD Student at ETH Zurich and Max Planck Institute. Focus on Computer Vision and 3D Human Modelling.
Vassilis Choutas @vchoutas1
2K Followers 2K Following Research Scientist @Google, Ph.D. from @PerceivingSys and @ETH, prev. intern @Microsoft and @RealityLabs, ECE @Aristoteleio, trying to capture 3D humans
Timo Bolkart @BolkartTimo
2K Followers 496 Following Research Scientist at Google, previously Research Scientist at Max Planck Institute (@MPI_IS) and Visiting Academic @Amazon.
Hongyi Wang @HongyiWang10
1K Followers 1K Following Senior Project Scientist @mldcmu @CarnegieMellon; MLSys researcher; Member @llm360; Ph.D. @WisconsinCS; On the academic job market NOW!
Simo Ryu @cloneofsimo
3K Followers 383 Following #KAIST RAI Lab (ML engineering #Naver) Interested in robotics, RL, math (but you might know me for t2i diffusion) [email protected]
Yongchang Hao @yongchanghao
158 Followers 649 Following PhD student @UAlbertaCS w/ @AmiiThinks. Ex-intern @TencentGlobal AI Lab and @Google.
Brian Cheung @thisismyhat
4K Followers 548 Following This is my hat, there are many like it, but this one is mine. @MIT_CSAIL 🧢 | @berkeley_ai 🎓 | Google B̶r̶a̶i̶n̶ DeepMind 🎩
PicoCreator (🇸🇬.. @picocreator
2K Followers 162 Following Builds Attention-Free Transformer (https://t.co/YL7CbNYKBs) from scratch - CEO @ https://t.co/kQHiGtzJWr Also built k8s tools, uilicious & GPU.js (https://t.co/OIfnI1EPU7)
BlinkDL @BlinkDL_AI
7K Followers 90 Following RWKV = 100% RNN with GPT-level performance. https://t.co/TkdxOJSFWX and https://t.co/86DzS6arA0
Ethan @Ethan_smith_20
3K Followers 685 Following a boy and his gpu vs the world. directing research at @leonardoai_. learning as I go. uf psych. generative models and representation learning
Itamar Zimerman @ItamarZimerman
253 Followers 330 Following PhD candidate @ Tel Aviv University. AI Research scientist @ IBM Research. Interested in deep learning and algorithms.
Historic Vids @historyinmemes
5.1M Followers 208 Following Daily history lessons. Education through memes!
Samuel L Smith @SamuelMLSmith
2K Followers 361 Following Research Scientist at DeepMind. Optimization and Initialization. Formerly Google Brain. Ex-Physicist.
UiPath @UiPath
104K Followers 5K Following We envision a world with a 🤖 for every person. Dedicated to accelerating human achievement via an #AI-powered end-to-end #automation platform.
UCL Centre for Artifi.. @ai_ucl
9K Followers 53 Following Officially Launched in September 2019, we are the home of Artificial Intelligence research and study at UCL📢📢Most diffusion (and flow matching) models use handcrafted schedules for their denoising steps during sampling. We show how to optimize them in a principled manner for high-quality generation! @amsabour added quickstart guide & collab to get you started quickly (links below)!
📢📢 Align Your Steps: Optimizing Sampling Schedules in Diffusion Models research.nvidia.com/labs/toronto-a… TL;DR: We introduce a method for obtaining improved sampling schedules for diffusion models, resulting in better samples at the same computation cost. (1/5)
It's been almost a year since I defended my PhD... But I haven't talked much about my Ph.D. research here... I haven't even shared some of my final Ph.D. papers about applying AI to microscopy and pathology... If people are interested, will share some threads about it soon
product managers in eng meetings
a video by @eigensteve suggested that a system like a pendulum that can be described with a single parameter (angle), we could potentially learn an autoencoder with only a single latent variable and it could recover that role. after 5k steps it looks like it did pretty solid
My favorite part is that it works really well with out-of-the-distribution garments
Testing out the new virtual try-on pipeline on @huggingface, IDM-VTON ▶️ huggingface.co/spaces/yisol/I…
last thursday, Meta dropped Llama 3, the OpenAI killer. no doubt a very impressive model! but over the weekend, we discovered an extremely trivial programmatic jailbreak against llama 3...sorry zuck!😘 so much for all that safety-tuning☹️ code: github.com/haizelabs/llam…
🆕 Introducing JAT, the first open-source multi-modal, multi-task multi-domain agent! 🤖 A step toward open generalist agents! 🚀 📰 Blog: huggingface.co/blog/jat
@_Tobie__ There's a variable there that is an unknown and linked to data quality, so, yes
@Teknium1 Wonder if this complies with chinchilla "optimal compute" training recipes
Let's give er' a go!
Microsoft just released Phi-3 - phi-3-mini: 3.8B model trained on 3.3T tokens rivals Mixtral 8x7B and GPT-3.5 - phi-3-medium: 14B model trained on 4.8T tokens w/ 78% on MMLU and 8.9 on MT-bench arxiv.org/abs/2404.14219
Microsoft just released Phi-3 - phi-3-mini: 3.8B model trained on 3.3T tokens rivals Mixtral 8x7B and GPT-3.5 - phi-3-medium: 14B model trained on 4.8T tokens w/ 78% on MMLU and 8.9 on MT-bench arxiv.org/abs/2404.14219
Llama-3-8b already dethroned? The benchmarks look really good! Their 7b model is apparently significantly better Llama-3-8b!! 👀👀 Really excited to try this model out, hope it gets released soon!
Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone abs: arxiv.org/abs/2404.14219 Microsoft announces phi-3-mini, a 3.8B model trained on 3.3T tokens that rivals Mixtral 8x7B and GPT-3.5 Has same arch as Llama-2 to benefit open-source community Also…
5T tokens FineWeb dataset just dropped @huggingface It's a 275GB dataset with cleaned and deduplicated data under an Open Data Commons license. We all see the difference the 15T tokens pre-training made for LLaMA-3 and now everyone can have it .
Llama-3 is absolutely impressive, but is it more resilient to adaptive jailbreak attacks compared to Llama-2? 🤔 Not much. The same approach as in our recent work arxiv.org/abs/2404.02151 leads to 100% attack success rate. The code and logs of the attack are now available:…
15T tokens DataLoader, you're welcome
We have just released 🍷 FineWeb: 15 trillion tokens of high quality web data. We filtered and deduplicated all CommonCrawl between 2013 and 2024. Models trained on FineWeb outperform RefinedWeb, C4, DolmaV1.6, The Pile and SlimPajama!
Zuck releasing a billion dollar model is actually wild, like really undermining what OAI is doing. flexing compute like “yea we can do that not a big deal”
Trends for United States
You might like
