eddy @eddy_data3
Working on reasoning and synthetic data Joined May 2022-
Tweets479
-
Followers459
-
Following5K
-
Likes8K
The race for LLM "cognitive core" - a few billion param model that maximally sacrifices encyclopedic knowledge for capability. It lives always-on and by default on every computer as the kernel of LLM personal computing. Its features are slowly crystalizing: - Natively multimodal…
The race for LLM "cognitive core" - a few billion param model that maximally sacrifices encyclopedic knowledge for capability. It lives always-on and by default on every computer as the kernel of LLM personal computing. Its features are slowly crystalizing: - Natively multimodal…
We reproduced DeepSeek R1-Zero in the CountDown game, and it just works Through RL, the 3B base LM develops self-verification and search abilities all on its own You can experience the Ahah moment yourself for < $30 Code: github.com/Jiayi-Pan/Tiny… Here's what we learned 🧵
Introducing 🔥Aria-Chat🔥, our latest multimodal chat model optimized for open-ended and multi-round dialogs! It outperforms Aria by 7 points on WildVision-Bench, offering enhanced reliability and stronger multilingual support. Download the model now: huggingface.co/rhymes-ai/Aria…
🚨 Our latest work evaluates the synthetic data generation abilities of different LLMs. Key findings: - Strong task-solvers ≠ strong synthetic data generators. - Bigger isn’t always better! E.g., Llama3.1-8B can outperform a 405B model in certain settings. - More synthetic data…
🚨 Our latest work evaluates the synthetic data generation abilities of different LLMs. Key findings: - Strong task-solvers ≠ strong synthetic data generators. - Bigger isn’t always better! E.g., Llama3.1-8B can outperform a 405B model in certain settings. - More synthetic data… https://t.co/IPSKVJAssa
We’ve been busy building an AI system to generate 3D worlds from a single image. Check out some early results on our site, where you can interact with our scenes directly in the browser! worldlabs.ai/blog 1/n
New Model From Rhymes! ✨ We're thrilled to announce, Allegro — a small and efficient open-source text-to-video model that transforms your text into stunning 6-second videos at 15 FPS and 720p! 🚀 🩷 Explore Allegro - Gallery: rhymes.ai/allegro_gallery - @huggingface:…
🚀 Introducing Aria from @rhymes_ai_ : The first open-source, multimodal native MoE model! Aria uses 3.9B parameters per token, excelling in multimodal & language tasks. It features a 64K token context window, captioning 256-frame videos in 10 seconds. Lightweight, fast, &…
prooftech is fundamentally anarchist
the cognitive profile of humans is not generally distributed across all cognitive skills. for example we’re unreasonably good at modeling our friends theory of mind and super bad at learning ten languages or difficult math (except on the margins)
🚀 Introducing MMMU-Pro: A more robust version of MMMU arxiv.org/pdf/2409.02813… After launching MMMU, we received valuable feedback from the community: 1️⃣ Some questions were answerable without even seeing the images. 2️⃣ Models didn’t always "know" the answer but found shortcuts…
What if you could use all the computing power in the world to train a shared, open source AI model? Preliminary report: github.com/NousResearch/D… Nous Research is proud to release a preliminary report on DisTrO (Distributed Training Over-the-Internet) a family of…
It’s Sunday morning we have some time with the coffee so let me tell you about some of our recent surprising journey in synthetic data and small language models. This post is prompted by the coming release of an instant, in-browser model called SmolLM360 (link at the end) The…
LLM model size competition is intensifying… backwards! My bet is that we'll see models that "think" very well and reliably that are very very small. There is most likely a setting even of GPT-2 parameters for which most people will consider GPT-2 "smart". The reason current…
LLM model size competition is intensifying… backwards! My bet is that we'll see models that "think" very well and reliably that are very very small. There is most likely a setting even of GPT-2 parameters for which most people will consider GPT-2 "smart". The reason current…
> Obstinacy is a reflexive resistance to changing one's ideas. If you have a model M of the world which predicts X with confidence X', and you observe ~X, then: M -> X = ~M or X = ~M The most obstinate may reject the evidence for ~X. It's usually noisy. Maybe you made a…
> Obstinacy is a reflexive resistance to changing one's ideas. If you have a model M of the world which predicts X with confidence X', and you observe ~X, then: M -> X = ~M or X = ~M The most obstinate may reject the evidence for ~X. It's usually noisy. Maybe you made a…
@_xjdr Thinking about emergence, feels that we can semi-rigorously say which tasks are hard-locked by serial operation depth or by width/fundamental richness limits on LLM representations, and can only be approached asymptotically with compute in small models; and which are… just hard.
Love @ylecun ‘s spicy takes 😂 I see his point though — is gradient descent “learning” if you limit it to a few iterations? The life expectancy of a human is short in the timescale of technological progress. Science has a scaling law too 🤔
Love @ylecun ‘s spicy takes 😂 I see his point though — is gradient descent “learning” if you limit it to a few iterations? The life expectancy of a human is short in the timescale of technological progress. Science has a scaling law too 🤔
Coool
Thank @_akhaliq for sharing our work! Paper: arxiv.org/pdf/2405.15071 Key takeaways: 1) Transformers can learn to implicitly reason, but only through extended training far beyond overfitting, a phenomenon known as grokking. 2) Transformers exhibit different levels of…
Thank @_akhaliq for sharing our work! Paper: arxiv.org/pdf/2405.15071 Key takeaways: 1) Transformers can learn to implicitly reason, but only through extended training far beyond overfitting, a phenomenon known as grokking. 2) Transformers exhibit different levels of… https://t.co/63d6MJ25TS
01.AI Yi-Large-Preview ranks only 2 places under the newest @OpenAI @ChatGPTapp GPT-4o on @alpacaml Eval2.0 verified category. Not bad, we'd say. Your thought? 😉 tatsu-lab.github.io/alpaca_eval ➡️ Yi-Large API beta sign up: platform.01.ai
7Layer | Overclock Va... @7LayerMagik
7K Followers 6K Following Be good to the core(s). @OverclockSol validator
jac0xb.sol @jacobdotsol
10K Followers 3K Following @TipLinkOfficial engineer | @LighthouseProto contributor | prev: @solsniperxyz | prev: @MonkeDAO board member
Rex @R89Capital
71K Followers 5K Following
Stepan | squads.xyz @SimkinStepan
9K Followers 3K Following CEO @SquadsProtocol. Building stablecoin accounts for businesses (@tryaltitude) and consumers (@fusewallet).
Darcy @coates_darcy51
265 Followers 3K Following
RenataDefoe @354D1T55SJwNSv
0 Followers 82 Following
PrimaBarrie @2oqbZ3Rj3iiK54
17 Followers 1K Following
AI Native Foundation @AINativeF
4K Followers 5K Following Non-profit Org., Empowering Humanity with Ethical AI, Latest insights about AI Native. 🤝 Community: https://t.co/b1mRBfQYi5
TiffanyClare @hNA32oSvdu35bLv
0 Followers 118 Following
ValerieBart @32u9xwl087RXQ0K
4 Followers 456 Following
Evan Barrows @barrows_ev95201
57 Followers 3K Following
SelenaGregory @Xb60rDk37J2H4Y1
44 Followers 2K Following
DinahNewman @3dCQ41Q2f0Ed1r6
20 Followers 2K Following
Lakshya A Agrawal @LakshyAAAgrawal
2K Followers 2K Following AI PhD @ UC Berkeley | GEPA Creator (https://t.co/EdPqvzj7k4) | Created https://t.co/YxPZsXZJeS | Past: AI4Code Research Fellow @MSFTResearch | Hobbyist Saxophonist
JillNewton @3CpE5HSzKpZmSv
11 Followers 700 Following
StephanieJob @5QtyU0V91m6NW7
18 Followers 658 Following
PrudenceWard @X5qd8486MExf97
15 Followers 436 Following
DarkPoolScans🇺🇸 @Toaja672370
41 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Savannah @8Wu9qs668C4ysU3
26 Followers 1K Following
Uipupa @Uipupa702
34 Followers 1K Following
Qawjou @Qawjou8533931
69 Followers 2K Following
Mavis Wanczyk @StaceyJenn68372
230 Followers 6K Following I’m Mavis Wanckzy the Powerball jackpot lottery winners of $758 million in MA Giving out $5,000 to 300 individual.
Eplawxal @Eplawxal141
34 Followers 1K Following
Yuta @2501KAMAMESI
39 Followers 1K Following 最近、高校生の頃に怖くて挫折した「ガンツ」を頑張って読破しています。どうやら自律神経に疾患があるようで好不調に波があります。
Fawe @Fawe3995519
89 Followers 2K Following
SwingChartist🇺🇸 @Awkawtji272364
38 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
はたあんじ @hataanji5748
92 Followers 3K Following
Selena Nolan @selena_nol2095
73 Followers 1K Following
GlobalMacroX🇺🇸 @Aveagau318570
39 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Maxwell Kerluke @MKerluke92491
29 Followers 2K Following
Nawpreh @Nawpreh391
22 Followers 713 Following
cherry @cherry19980809
58 Followers 1K Following
Rohan Paul @rohanpaul_ai
83K Followers 8K Following Compiling in real-time, the race towards AGI. 🗞️ Don't miss my daily top 1% AI analysis newsletter directly to your inbox 👉 https://t.co/6LBxO8215l
SupportResist🇺🇸 @Iecooogal27138
40 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Louie Kautzer @LKautzer79222
18 Followers 2K Following
Sigrid Hoeger @HoegerSigr56751
31 Followers 2K Following
Qumo @Qumo624
33 Followers 1K Following
EarningsSurprise🇺�... @Yxausvoo1709
44 Followers 2K Following 15-30% Monthly | 2 High-Conviction Stocks.Short-Term Gains: 15-20% in Days/Weeks.DM "JOIN" for WhatsApp Alerts. Live Trade Signals • Market Analysis
Anabel Lockman @AnabelLock66471
74 Followers 3K Following
Longtao Zheng @ltzheng01
150 Followers 570 Following PhD student @NTUsg. Training open-ended agents in open-ended worlds
Ken Osinski-Goyette @GoyetteKen45098
88 Followers 3K Following
Dell Kihn @DellKihn94032
82 Followers 4K Following
Yang Chen @ychenNLP
1K Followers 554 Following Research Scientist @NVIDIA | PhD @GeorgiaTech| RL and LLM reasoning
SORA @habatakusora
8 Followers 393 Following
important paper @moscowluvextasy
195 Followers 2K Following 🇷🇺 | We know almost nothing about important paper, but we are sure important paper is a great person.
joseph greene @jjgreeneUSA
1K Followers 3K Following I've invested my entire networth (10907 TJS) into the next bitcoin of technology AIDA
toly 🇺🇸 @aeyakovenko
628K Followers 6K Following Co-Founder of Solana Labs. Award winning phone creator. NFA, don’t trust me, mostly technical gibberish. https://t.co/LomgbTpb6h
Solana @solana
3.5M Followers 3K Following The high performance network powering internet capital markets, payments, and crypto applications. ◎ X By @SolanaFndn
Noah 🎈 @redacted_noah
19K Followers 2K Following Protocol engineering at @helium. Opinions are my own.
raj 🖤 @rajgokal
337K Followers 2K Following @solana accelerationist. giga-techno-optimist. bad jokes are my own. rt's and likes are not endorsements. do your own research.
Austin Federa | 🇺�... @Austin_Federa
163K Followers 2K Following cofounder of @doublezero | increase bandwidth reduce latency | be kind | fmr strategy @solana | Everything's Computer
Armani Ferrante @armaniferrante
147K Followers 4K Following Founder, CEO @Backpack @MadLads @anchorlang. Jobs: https://t.co/D6Fs0yprr1 Exchange: https://t.co/GdGbaa0D2n App: https://t.co/UR4AAqAO25
BONK!!! @bonk_inu
456K Followers 377 Following The community token of web3 | Token ID: DezXAZ8z7PnrnRJjz3wXBoRgixCa6xjnB7YaB1pPB263
Solana Legend 🎒 @SolanaLegend
200K Followers 4K Following Co-Founder, Managing Partner @FrictionlessVC, @monkeDAO Opinions are my own, not investment advice
mert | helius.dev @0xMert_
255K Followers 2K Following ceo @heliuslabs, ex @coinbase — Solana RPCs, APIs, trading infra: https://t.co/J2To2Mz9GF — the best crypto pod: https://t.co/87uPmEFrEL
7Layer | Overclock Va... @7LayerMagik
7K Followers 6K Following Be good to the core(s). @OverclockSol validator
jac0xb.sol @jacobdotsol
10K Followers 3K Following @TipLinkOfficial engineer | @LighthouseProto contributor | prev: @solsniperxyz | prev: @MonkeDAO board member
DRiP 💧 @drip_haus
240K Followers 1K Following The Collectibles Platform. Collect, create, and earn on crypto's #1 content app. Corporate: @drip
Jordan ◎ | 🥫🪽 @jordaaash
24K Followers 2K Following Engineer @anza_xyz / Advisor @solanalabs / Collector @MadLads
0xngmi @0xngmi
172K Followers 282 Following Building a transparent defi dashboard at @defillama. Code llama @llamapay_io @sealed_art @smolrefuel Would rather be right than make money
Rex @R89Capital
71K Followers 5K Following
Jacob Creech @jacobvcreech
20K Followers 1K Following @SolanaFndn Head of Developer Relations · Developer · Contribute to Solana Ecosystem https://t.co/4DYZWFnEY8
Ryan Selkis (d/acc) �... @Selkis_2028
379K Followers 2K Following I am @Solomon_2028 | American Supremacist & Girardian | OFFENSE
Laine ❤️ by SOL S... @laine_sa_
21K Followers 191 Following We operate a leading validator on Solana as well as https://t.co/4yRdhRpekB. Part of @solstrategies_
James @jamesrp13
4K Followers 964 Following Building Bookloom so you can get more from the books you love. Founder @unboxedsoftware where we turn raw ideas into launch-ready products. Co-host @SolfatePod.
Yujia Qin @TsingYoga
5K Followers 331 Following ByteDancer, Agent, THU (16-20 BS in EE, 20-24 PhD in CS)
Towaki Takikawa / 瀧... @yongyuanxi
9K Followers 3K Following ceo @outerport (yc s24), former research scientist @nvidia, cs @uwaterloo, phdish @uoft, making technical documents and cad AI-readable
Z.ai @Zai_org
15K Followers 142 Following The AI lab behind GLM models, dedicated to inspiring the development of AGI to benefit humanity. https://t.co/b6zGxJvzzS
Ari Morcos @arimorcos
7K Followers 2K Following CEO and Co-founder @datologyai working to make it easy for anyone to make the most of their data. Former: RS @AIatMeta (FAIR), RS @DeepMind, PhD @PiN_Harvard.
Jacob Austin @jacobaustin132
7K Followers 917 Following Research at @GoogleDeepMind. Currently making LLMs go fast. I also play piano and climb. NYC. Opinions my own
Ihtesham Haider @ihteshamit
160K Followers 551 Following co-founder of @theprohumanai, running a brand agency & ai businesses, sharing what i learn here so you can be more productive and successful.
Vaish Shrivastava @VaishShrivas
358 Followers 417 Following Reinforcement Learning @MSFTResearch 🧠 MS CS @Stanford @StanfordNLP @StanfordAILab BS CS @Caltech
Alex Trott @alexrtrott
786 Followers 289 Following Research @DbrxMosaicAI. Neuroscience PhD in a previous life. Whispering models into sentience one parameter at a time. (opinions are my own.)
Lukasz Olejnik @lukOlejnik
18K Followers 270 Following Security & Privacy. Data Protection. Research. Engineering. Analyst. Policy. W3C. Consultant. Author. Ph.D, LL.M. @warstudies
bone @boneGPT
41K Followers 7K Following talentless hack | https://t.co/b6gnaCp4C6 | https://t.co/ptcWUTW8Jl | bonecoin
RL_Conference @RL_Conference
5K Followers 256 Following Home of the first annual reinforcement learning conference. Stay tuned up for updates!
Jiaxuan You @youjiaxuan
3K Followers 209 Following Assistant Professor at @UIUC CS, directing @ULab @Stanford CS PhD Senior Research Scientist @NVIDIA #graph #LLM #Agent #GenAI
Lakshya A Agrawal @LakshyAAAgrawal
2K Followers 2K Following AI PhD @ UC Berkeley | GEPA Creator (https://t.co/EdPqvzj7k4) | Created https://t.co/YxPZsXZJeS | Past: AI4Code Research Fellow @MSFTResearch | Hobbyist Saxophonist
Guan Wang @makingAGI
5K Followers 36 Following CEO of Sapient Intelligence. Exploring the path to AGI through brain-inspired AI. 🧠🤖 #AGI #NeuroAI
Yuxi on the Wired @layer07_yuxi
3K Followers 71 Following [email protected]:~$ whoami tech enby (agi/asi pronouns) (math ⊗ physic ⊗ comput).ist
Jamie Voynow @jamievoynow
6K Followers 593 Following Building AI agents for quant finance @ https://t.co/VSMWf23KZo
Aparna Dhinakaran @aparnadhinak
6K Followers 1K Following AI Founder: building @arizeai & @arizephoenix 💙 I post about LLMs, LLMOps, Generative AI, ML and occasionally Amazing Race
Zeyu Huang @ZeroyuHuang
130 Followers 129 Following PhD @EdinburghNLP Working on LLM | Student Researcher @GoogleDeepmind
Indra @IndraVahan
12K Followers 669 Following lachine mearning engineepreneur who says silly things, ai guy @aphelionlabs
Mind Essentials @Mind_Essentials
1.3M Followers 22 Following Self-Discipline | Philosophy Thoughts | Psychology Hacks | Wealth & Wisdom | Acceptance | Awareness
Thomas Capelle @capetorch
4K Followers 1K Following Chilean 🇨🇱 living in France. I build DL models and pipelines. ML Engineer at @weights_biases cargobike ♥🚴
Longtao Zheng @ltzheng01
150 Followers 570 Following PhD student @NTUsg. Training open-ended agents in open-ended worlds
Tyler Griggs @tyler_griggs_
553 Followers 349 Following CS PhD student @UCBerkeley Sky Lab, co-leading @NovaSkyAI and building SkyRL | Previously @GoogleCloud infra | @Harvard 2020
Olexandr Isayev 🇺�... @olexandr
10K Followers 4K Following #CompChem Prof @CMU_Chem. Connecting chemical sciences with AI & #MachineLearning. #tarheels fan. Care:#design, #photography #Ukraine #cats🐈 Rants are mine
Yaashaa Golovanov @Golovanov_ammoc
10K Followers 584 Following Director, AMMOC - An International Math Circle in honor of Vladimir Igorevich Arnold and Jerrold Eldon Marsden.
Arshdeep Singh @arsh99_singh
5K Followers 269 Following engineer @xAI | ex: @Relicx @Google @bitspilaniindia
Yoav Artzi @yoavartzi
17K Followers 183 Following Research/prof @cs_cornell + @cornell_tech🚡 / https://t.co/9YnWry7yHs / asso. faculty director @arxiv / building https://t.co/f9QkzO5kaC and @COLM_conf
Martin Klissarov @MartinKlissarov
2K Followers 695 Following AI Research @GoogleDeepMind, PhD in RL with Doina Precup and @MarlosCMachado. Previously @Apple & @Meta
Sarah Constantin @s_r_constantin
2K Followers 418 Following (im)practical utopian. married to @oscredwin.
SWE-bench @SWEbench
126 Followers 5 Following
Xiaohua Zhai @XiaohuaZhai
11K Followers 311 Following Researcher at Meta (previously at OpenAI Zürich, Google DeepMind)
Alexander Kolesnikov @__kolesnikov__
12K Followers 192 Following
Dave West @DRWestWing
599 Followers 112 Following Never stop working to advance learning around the world and close to home. [email protected]
Andy Konwinski @andykonwinski
8K Followers 98 Following Co-founder of @databricks, @perplexity_ai, @laudeinstitute, @laudeventures
Zhoujun (Jorge) Cheng @ChengZhoujun
989 Followers 594 Following CS Ph.D. @UCSanDiego | Prev. @XLangNLP @MSFTResearch @sjtu1896Trends for United States
You might like
