Brandon Amos @brandondamos
research scientist @MetaAI (FAIR) | optimization, machine learning, control, and reinforcement learning | PhD from @SCSatCMU bamos.github.io New York, NY Joined January 2014-
Tweets4K
-
Followers14K
-
Following2K
-
Likes8K
1/What does it mean for an LLM to “memorize” a doc? Exactly regurgitating a NYT article? Of course. Just training on NYT?Harder to say We take big strides in this discourse w/*Adversarial Compression* w/@A_v_i__S @zhilifeng @zacharylipton @zicokolter 🌐:locuslab.github.io/acr-memorizati…🧵
GenAI does not stop, so it is time for a new blog post. Since LLMs are everywhere, I decided to take a look at them, my curious readers. And a bonus: An implementation of a teenyGPT 🤖✨ 📄Post: jmtomczak.github.io/blog/20/20_llm… 🖥️Code: github.com/jmtomczak/intr…
Everything can be formulated as an optimization problem.
Everything can be formulated as an optimization problem.
New paper alert: React-OT: Optimal Transport for Generating Transition State in Chemical Reactions (arxiv.org/abs/2404.13430). React-OT formulates TS search as a transport problem, approaching chemical accuracy while taking only 0.5 seconds in inference on a single GPU. #compchem
We have a new preprint out - your language model is not a reward, it’s a Q function! 1. The likelihood of the preferred answer must go down - it’s a policy divergence 2. MCTS guided decoding on language is equivalent to likelihood search on DPO 3. DPO learns credit assignment
Excited to share Diffusion-DPO, a method to directly align diffusion models to user preference. DPO-tuned SDXL obtains a 70% win rate over SDXL on PartiPrompts, a new SOTA for open source models! It is also effective at Learning from AI Feedback. arxiv.org/abs/2311.12908 (1/N)
1/ New work on ML+PDEs: differentiable PDE-constrained optimization as a layer in neural networks can be made much faster by scaling up via mixture-of-experts → also better training stability and improved accuracy! Accepted at #ICLR2024: openreview.net/forum?id=u3dX2…
same with my paper collection 😅 (except instead of buying them I pay for cloud storage so I can save as many vision and graphics papers as I want in full resolution)
same with my paper collection 😅 (except instead of buying them I pay for cloud storage so I can save as many vision and graphics papers as I want in full resolution) https://t.co/vpBNZbc8Gg
We studied In-Context learning with hundreds to thousands of examples. My favorite example: I sent *one million* tokens to Gemini 1.5 Pro for linear classification with 64 dimensional integer-valued vectors and many-shot learning performs similarly to k-Nearest Neighbours.
New #NVIDIA paper: Real-time text-to-3D generation #ICCV2023 3D generation from text requires expensive per-prompt optimization. We train 1 model on many prompts for real-time generalization to unseen prompts, interpolations and more! ATT3D details: research.nvidia.com/labs/toronto-a…
Our team at PNNL is hiring passionate postdoc to do research in Scientific Machine Learning methods with applications to applied energy and beyond. careers.pnnl.gov/jobs/9003?lang…
I'm delighted to announce a differentiable INLA implementation in JAX, from my colleague @geraschenko. In principle this lets you fit latent Gaussian MRFs (e.g., for spatial stats, or SLAM) with gradient-based methods, although currently the memory use is prohibitive . The trick…
Introducing 𝐀𝐋𝐎𝐇𝐀 𝐔𝐧𝐥𝐞𝐚𝐬𝐡𝐞𝐝 🌋 - Pushing the boundaries of dexterity with low-cost robots and AI. @GoogleDeepMind Finally got to share some videos after a few months. Robots are fully autonomous filmed in one continuous shot. Enjoy!
🚀 How can meta-learning, self-attention & JAX power the next generation of Evolutionary Optimizers 🦎? Excited to share my @DeepMind internship project and our #ICLR2023 paper ‘Discovering Evolution Strategies via Meta-Black-Box Optimization’ 🎉 📜: openreview.net/forum?id=mFDU0…
I’m super excited to release our 100+ page collaborative agenda - led by @usmananwar391 - on “Foundational Challenges In Assuring Alignment and Safety of LLMs” alongside 35+ co-authors from NLP, ML, and AI Safety communities! Some highlights below...
🦎Can we teach Transformers to perform in-context Evolutionary Optimization? Surely! We propose Evolutionary Algorithm Distillation for pre-training Transformers to mimic teachers 🧑🏫 🎉 Work done @GoogleDeepMind 🗼with @alanyttian & @yujin_tang 🤗 📜: arxiv.org/abs/2403.02985
5 more days to apply for a summer internship in Scientific Machine Learning at PNNL! careers.pnnl.gov/jobs/8943?lang…
🎨Spent some time refactoring the 2021 post on diffusion model with new content: lilianweng.github.io/posts/2021-07-… ⬇️ ⬇️ ⬇️ 🎬Then another short piece on diffusion video models: lilianweng.github.io/posts/2024-04-… (Yes, I had an intensive weekend🥹)
There's no reason not to use resets to speed up RL in LLM-land, where it's just generating from a prefix! You can also use the broader set of hybrid RL techniques to speed up *repeated* RL (e.g. in inverse RL --gokul.dev/filter/ or arxiv.org/abs/2402.08848).
There's no reason not to use resets to speed up RL in LLM-land, where it's just generating from a prefix! You can also use the broader set of hybrid RL techniques to speed up *repeated* RL (e.g. in inverse RL --gokul.dev/filter/ or arxiv.org/abs/2402.08848).
Differentiable Metropolis-Hastings: differentiate through Bayesian estimation to optimize models towards achieving desired probabilistic outcomes, with implementation in #julialang (#sciml) For more information, see arxiv.org/abs/2306.07961
Soumith Chintala @soumithchintala
186K 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.Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0pKevin Patrick Murphy @sirbayes
42K Followers 334 Following Research Scientist at Google Brain / Deepmind. Interested in Bayesian Machine Learning.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).Eugene Vinitsky @EugeneVinitsky
13K Followers 2K Following Anti-cynic. Artificial narrow intelligence. Autonomous vehicles, multi-agent learning, and transportation. RS at Apple, Asst. Prof at @nyutandon. He/him.Sergey Levine @svlevine
79K Followers 122 Following Associate Professor at UC Berkeley Co-founder, Physical IntelligenceDan Roy @roydanroy
45K Followers 2K Following ML / AI researcher, emphasis on theory. Research Director and Canada CIFAR AI Chair, @VectorInst Professor, @UofT (Statistics/CS)Noam Brown @polynoamial
34K Followers 611 Following Researching reasoning @OpenAI | Co-created Libratus/Pluribus, the first superhuman no-limit poker AIs | Co-created CICERO | PhD from @SCSatCMUHorace He @cHHillee
23K Followers 449 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemaleyobibyte @y0b1byte
15K Followers 2K Following Kurin ViTaly, senior research scientist @IsomorphicLabs, ML PhD from @UniofOxford on RL, Multitask learning & GraphsEdward Grefenstette @egrefen
36K Followers 774 Following FR/US/GB AI/ML Person, Director of Research at @GoogleDeepMind, Honorary Professor at @UCL_DARK, @ELLISforEurope Fellow. All posts are personal.Karol Hausman @hausman_k
22K Followers 141 Following @Physical_int ex: researcher @GoogleAI/@DeepMind, adj. Prof. @Stanford. Into robots, AI, NBA, philosophy, soccer and almond croissants. 🇵🇱🇺🇸Sander Dieleman @sedielem
50K Followers 2K Following Research Scientist at Google DeepMind. I tweet about deep learning (research + software), music, generative models (personal account).Natasha Jaques @natashajaques
25K Followers 1K Following Senior Research Scientist at @GoogleAI and Assistant Professor @uwcse. Social Reinforcement Learning in multi-agent and human-AI interactions. PhD from @MIT.Prof. Anima Anandkuma.. @AnimaAnandkumar
25K Followers 2K Following Bren Professor @caltech, Fmr Sr Director of #AI research @nvidia, Fmr Principal Scientist @awscloud, AI+Science, PDE, Neural operators. Views my own.Danijar Hafner @danijarh
14K Followers 868 Following Building AI that makes autonomous decisions using world models, artificial curiosity, and temporal abstraction @DeepMindSasha Rush @srush_nlp
52K Followers 464 Following Professor, Programmer in NYC. Cornell Tech, Hugging Face 🤗 https://t.co/cZl0wTfqGzMichael Bronstein @mmbronstein
43K Followers 4K Following #DeepMind Professor of #AI @UniofOxford / Fellow @ExeterCollegeOx / ML Lead @ProjectCETI / https://t.co/kZpGpDzYeVAnimesh Garg @animesh_garg
21K Followers 1K Following Foundation Models for Generalizable Autonomy. Assistant Professor in AI Robotics @GeorgiaTech + @NvidiaAI. prev @Stanford @berkeley_ai @UofTCompSciL G 🇺🇸 @LGSouzaB
216 Followers 2K Following Lawyer, tech investor, software engineering, space enthusiast and geopolitics aficionado. Invigorated by an exhilarating dance with RISK. Crypto since 2012.Avi Schwarzschild @A_v_i__S
267 Followers 183 Following Postdoc at CMU. Trying to learn about deep learning faster than deep learning can learn about me.Sakuye Entertainer�.. @SakuyeEnte16474
235 Followers 1K Following Use my promo-code ''SAKUYE'' to register on 1xbet & you'll get 300% bonus on your first deposit, goodluck.Sandee Ardelia @SandeeArde54917
0 Followers 1 FollowingInwoo Hwang @InwooRyanHwang
117 Followers 490 Following PhD student @ Seoul National University, w/@sanghack. Interested in the intersection of causality and machine learning. Currently looking for internship in 2024Qq @zwu9048
2 Followers 90 FollowingBruno @bgmirand_
20 Followers 59 FollowingYibo Yang @YiboYang
220 Followers 101 Following PhD student at UC Irvine, working on machine learning + info theory + compression + LLMs. On the job market for postdoc & industry positions.Carl Grafe @CarlGrafe
925 Followers 1K Following Data analyst / consultant / problem solver @byuidaho | informatics PhD | epidemiology MS | sims | machine learning | math.Jud @Jud10427603
2 Followers 30 FollowingBeef with Big Data @beefwithbigdata
27 Followers 53 Following Exploring ideas in data analytics to optimize efficiency and sustainability in the beef industry. Please visit my blog at the link below.Pratyush Maini @pratyushmaini
1K Followers 340 Following Trustworthy ML | PhD student @mldcmu | Founding Member @datologyai | Prev. Comp Sc @iitdelhithenormalone @AkNiloy6
352 Followers 4K Following ML Engineer | Researcher | Musician | Lifelong Liverpool Fan 🇧🇩Nick Mumero @nickdee96
131 Followers 1K Following Cofounder at Continuum Ads. Focusing on NLP, Simulation Modelling and Optimization.Jeff Tatarchuk @jtatarchuk
1K Followers 2K Following Co-founder @tensorwavecloud - Pioneering the next wave of AI compute. Need GPUs? DM me.alan as a swe @asasoftware
149 Followers 1K Following Python and machine learning enthusiast. Currently studying software engineering and exploring the latest memes in tech.TSH @tas046
77 Followers 202 Followingsweetcarrot @zauso2
27 Followers 148 FollowingSafara @safara_travels
2K Followers 785 Following Safara curates the world's best hotels and rewards you with cashback on every booking.RM @desi_tweet
83 Followers 2K Followingים איתן @ytn_ym
1 Followers 53 FollowingBhavin @0xbhavin
61 Followers 370 Following Building something in AI. Previously: @SkySQL @Kobai_Inc @velocity_devMark J. @MarkJ97517270
55 Followers 246 FollowingGagan @encrypted_soul_
187 Followers 801 Following data and engineering @blaze_ai | crypto since 2023ByeRose @byerose365
0 Followers 520 FollowingMichael Molin @thematrixcom
1K Followers 2K Following General Intelligence System - https://t.co/56AXhKlqhq - Deep explorationChenru Duan @chenru_duan
764 Followers 452 Following ex-MSFT Quantum | Ph.D. @KulikGroup @MITChemistry | #AI4Science workshop organizer. #compchem, #MachineLearning, and #chemdiscovery.Raj Dave @dave_raj29
182 Followers 630 Following Management Consultant. LFC fan. Views and opinions are my own.Aditya Modi @adityamodi94
244 Followers 321 Following A theoretician hoping to apply RL in the wild world!El Capitano 🇧🇪�.. @El_Capitano_O
939 Followers 5K Following 🖥️ AI XR & Web3 Events Org.⚙️Engineer passionate about sustainable Health-Ed-Tech initiatives📚Book Club Owner and 🏋️♂️AthleteAIQUEST @ProAiquest
108 Followers 454 Following Exploring the latest in AI tools and technologies. Join me on a journey into the future of innovation and automation. #AI #chatgpt #TechEnthusiastAlbert Yu Sun (Going .. @Albertyusun
317 Followers 411 Following ML Research Engineer @dynamo_ai. 🔬: Privacy and NLP. Alum @DukeU. Former Research Intern: @MSFTResearch, @CuraiHQ, @ACLU, @VeraInstitute.Yutong (Kelly) He @electronickale
312 Followers 299 Following PhD student @mldcmu, I’m so delusional that doing generative modeling is my jobAbdul Manan | Power B.. @AbdullManaan
151 Followers 1K Following I help Businesses Make Data data-driven decisions by finding valuable insight using Power BI | Business Analytics Expert | Business AnalystYiding Jiang @yidingjiang
1K Followers 468 Following PhD student @mldcmu @SCSatCMU. Formerly intern @MetaAI, AI resident @GoogleAI. BS from @Berkeley_EECS. Trying to understand stuff.Steve Nordquist of C .. @CredentialedRSS
139 Followers 452 FollowingJianfeng Chi @jianfengchi
238 Followers 430 Following Research Scientist @AIatMeta, CS PhD @CS_UVA Opinions my ownJag @JagTangirala
16 Followers 417 Following Built a lot of routers that power today's Internet. Distinguished Engineer at Cisco.William @williamteo
58 Followers 239 FollowingGagan Jain @gaganjain1582
50 Followers 745 Following Predoc Researcher @GoogleDeepMind | IIT Bombay'22Soumith Chintala @soumithchintala
186K 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.Clément Canonne @ccanonne_
31K Followers 928 Following Senior Lecturer @Sydney_Uni. Postdocs @IBMResearch, @Stanford; PhD @Columbia. Converts ☕ into puns: sometimes theorems. He/him. @[email protected]Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0pKevin Patrick Murphy @sirbayes
42K Followers 334 Following Research Scientist at Google Brain / Deepmind. Interested in Bayesian Machine Learning.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).Lucas Beyer (bl16) @giffmana
56K Followers 444 Following Researcher (Google DeepMind/Brain in Zürich, ex-RWTH Aachen), Gamer, Hacker, Belgian. Mostly gave up trying mastodon as [email protected]Eugene Vinitsky @EugeneVinitsky
13K Followers 2K Following Anti-cynic. Artificial narrow intelligence. Autonomous vehicles, multi-agent learning, and transportation. RS at Apple, Asst. Prof at @nyutandon. He/him.Sergey Levine @svlevine
79K Followers 122 Following Associate Professor at UC Berkeley Co-founder, Physical IntelligenceDavid Pfau @pfau
22K Followers 1K Following Knowledge manifests itself in radiant dreams that shimmer like the wild sun Views are my own pfau at sigmoid dot social on 🦣 https://t.co/xqtVHHVI17 on 🦋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)Gautam Kamath @thegautamkamath
44K Followers 505 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Noam Brown @polynoamial
34K Followers 611 Following Researching reasoning @OpenAI | Co-created Libratus/Pluribus, the first superhuman no-limit poker AIs | Co-created CICERO | PhD from @SCSatCMUHorace He @cHHillee
23K Followers 449 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemaleyobibyte @y0b1byte
15K Followers 2K Following Kurin ViTaly, senior research scientist @IsomorphicLabs, ML PhD from @UniofOxford on RL, Multitask learning & GraphsPercy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistEdward Grefenstette @egrefen
36K Followers 774 Following FR/US/GB AI/ML Person, Director of Research at @GoogleDeepMind, Honorary Professor at @UCL_DARK, @ELLISforEurope Fellow. All posts are personal.Karol Hausman @hausman_k
22K Followers 141 Following @Physical_int ex: researcher @GoogleAI/@DeepMind, adj. Prof. @Stanford. Into robots, AI, NBA, philosophy, soccer and almond croissants. 🇵🇱🇺🇸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.Sander Dieleman @sedielem
50K Followers 2K Following Research Scientist at Google DeepMind. I tweet about deep learning (research + software), music, generative models (personal account).Yibo Yang @YiboYang
220 Followers 101 Following PhD student at UC Irvine, working on machine learning + info theory + compression + LLMs. On the job market for postdoc & industry positions.Pratyush Maini @pratyushmaini
1K Followers 340 Following Trustworthy ML | PhD student @mldcmu | Founding Member @datologyai | Prev. Comp Sc @iitdelhiChenru Duan @chenru_duan
764 Followers 452 Following ex-MSFT Quantum | Ph.D. @KulikGroup @MITChemistry | #AI4Science workshop organizer. #compchem, #MachineLearning, and #chemdiscovery.Yiding Jiang @yidingjiang
1K Followers 468 Following PhD student @mldcmu @SCSatCMU. Formerly intern @MetaAI, AI resident @GoogleAI. BS from @Berkeley_EECS. Trying to understand stuff.Adithya Murali @Adithya_Murali_
1K Followers 717 Following Research Scientist at @NVIDIAAI. Foundation models for robotics. Previously PhD at @CMU_Robotics, @Berkeley_EECS, @MetaAI, AWSAlex Li @alexlioralexli
632 Followers 344 Following PhD student in ML at @mldcmu. Prev: @AIatMeta and undergrad @berkeley_aiYutong (Kelly) He @electronickale
312 Followers 299 Following PhD student @mldcmu, I’m so delusional that doing generative modeling is my jobArsenii Ashukha @senya_ashuha
1K Followers 630 Following AI Research Scientist at @IsomorphicLabs reimagining drug discovery 💊🧬Yangqing Jia @jiayq
12K Followers 263 Following Founder @leptonai. @UCBerkeley alumni. ex @google & @facebook. ex vp @AlibabaGroup. Open source work on caffe, @pytorch, @tensorflow, & @onnxai.Tony Z. Zhao @tonyzzhao
12K Followers 780 Following CS PhD student @Stanford. Aspiring full-stack roboticist. Prev Deepmind, Tesla, GoogleX, Berkeley.Saurabh Garg @saurabh_garg67
863 Followers 579 Following Building next-gen AI at @MistralAI | prev/ PhD @mldcmu; CS @iitbombay (undergrad); Collab @GoogleAI @awscloud @appleTor Erlend Fjelde @torfjelde
407 Followers 146 Following PhD student in machine learning @CambridgeMLG @Cambridge_Uni. Contributor to @TuringLang. Mastodon: @[email protected]Daniel Dauner @DanielDauner
65 Followers 95 Following PhD student @AutoVisionGroup and @uni_tue working on autonomous drivingKashyap Chitta @kashyap7x
777 Followers 395 Following PhD @AutoVisionGroup @uni_tue on autonomous vehicles • Prev @CMU_Robotics @NVIDIAAI • @RSSPioneers 2023 • @CVPR @ICCVConference @NeurIPSConf '23 top reviewerStanley H. Chan @stanley_h_chan
7K Followers 137 Following Professor | computational imaging | machine learning | Purdue ECEErik Meijer @headinthebox
27K Followers 0 FollowingDongjun Kim @gimdong58085414
698 Followers 712 Following PostDoc at Stanford; Diffusion models; My own wordsDoron Haviv @DoronTheViking
242 Followers 364 Following PhD student @dana_peer lab. Formerly EE & Physics @TechnionLive. Disgruntled @SpursOfficial fan., Machine Learning, Spatial Transcriptomics.Fabian Schaipp @FSchaipp
457 Followers 502 Following PhD student in Optimization for Machine Learning at TU Munich.Historic Vids @historyinmemes
5.3M Followers 210 Following Daily history lessons. Education through memes!Omar Sanseviero @osanseviero
31K Followers 2K Following Chief Llama Officer @huggingface 🦙 Founder @AI_Learners. Xoogler (SWE @Google Assistant, 20% PM TF Graphics). 100% Hacker Llama🇵🇪🇲🇽Mengdi Wang @MengdiWang10
1K Followers 265 Following Princeton professor in AIML, optimization and data science. Program Chair @ICLR2023. Formerly @MIT @GoogleDeepmind @TsinghuaDurstewitzLab @DurstewitzLab
2K Followers 347 Following Scientific machine learning, AI & data analysis, dynamical systems theory, applications in (computat.) neuroscience & psychiatry. @[email protected]Kevin Stone @kevinleestone
368 Followers 273 Following Research @ OpenAI, previously at FAIR, TRI, and Google working on LLMs, RL, and Robotics.Jacob Helwig @JacobHelwig
276 Followers 749 Following TAMU CSCE, supervised by @ShuiwangJi. AI4ScienceRocky Duan @rocky_duan
778 Followers 84 Following Building @CovariantAI, CTO. Previously @OpenAI, @UCBerkeley PhD. 2024 Forbes 30 Under 30.KUNAL GARG @kunalgarg94
148 Followers 158 Following Postdoctoral associate at MIT, researching ML-based methods for safe Multiagent ControlHarri Edwards @HarriLEdwards
236 Followers 208 FollowingNayoung Jun @nayoung_jun
422 Followers 178 Following Research Scientist @Meta @RealityLabs; Ph.D. @DukeNeuroYingheng Wang @yingheng_wang
481 Followers 673 Following CS PhD Student @Cornell. Previously @JohnsHopkins, @Tsinghua_Uni, @MSFTResearch, @uwcse, @NECLabsAmerica.Bo Wang @BoWang87
8K Followers 2K Following Assistant Prof. CS,LMP @UofT; CIFAR AI Chair @VectorInst; Chief AI Scientist, @UHN; former PHD, CS @Stanford; opinions my own. #AI #healthcare #combioOmer Bar Tal @omerbartal
2K Followers 109 Following Founding Scientist @pika_labs | ex @WeizmannScience @GoogleAISherry Yang @mengjiao_yang
2K Followers 342 Following Research Scientist @GoogleDeepMind | PhD Student @UCBerkeley. Previously M.Eng. / B.S. @MIT.Ashley Edwards @ashrewards
485 Followers 200 Following Research scientist @GoogleDeepMind. Past: Uber AI Labs, Georgia TechYuge Shi (Jimmy) @YugeTen
4K Followers 476 Following 石宇歌 · Research Scientist @DeepMind · Past: PhD at Oxford, intern at Google Brain, FAIR, CSIRO · she/herXingyu Lin @Xingyu2017
1K Followers 335 Following Postdoc at @berkeley_ai. PhD from @SCSatCMU. #Learning #RoboticsJesse Farebrother @JesseFarebro
640 Followers 308 Following PhD student @Mila_Quebec / @McGillU. Student Researcher @GoogleDeepMind.Jane Dwivedi-Yu @JaneDwivedi
441 Followers 67 Following Researcher @MetaAI | Former PhD @UCBerkeley and @Cornell alumna.Tim Dettmers @Tim_Dettmers
29K Followers 820 Following PhD Student at @UW. I blog about deep learning and PhD life at https://t.co/Y78KDJJFE7.Lancelot Da Costa @lancelotdacosta
746 Followers 327 Following Researching the mathematics of intelligence 🧠👾 Maths, neuro & AI @ Imperial College, UCL & @VERSESAI Rarely on Twitter—contact me: [email protected]Raymond Chua @RaymondRChua
1K Followers 3K Following PhD @mcgillu and @Mila_Quebec. Into #AI 🤖 & #neuroscience 🧠. 🏊🏻🚴🏽♂️🏃🏻♂️🏕️ when away from 💻 .Hannah Lawrence @HLawrenceCS
517 Followers 398 Following PhD @ MIT CSAIL. Geometric deep learning, especially learning with symmetries (equivariance). https://t.co/0XcSE5V8S2Stephan Mandt @StephanMandt
2K Followers 556 Following ML Professor @UCIrvine, previously @blei_lab, @Princeton. #GenerativeAI, #Compression, #AI4Science. Program Chair @aistats_conf 2024; General Chair AISTATS 2025Two students came in for my office hours. […] Six hours later I took them to have ice-cream 🍦🍦🍦 🥲🥲🥲
Excited to be joining friends and colleagues from @GoogleDeepMind in Vienna for #ICLR2024 in a little over a week! Looking forward to meeting new people and hearing about exciting work on open-ended LLM-powered agents, tool use, and many other topics close to my heart ☺️
Who's building a terminal with LLM completion? Or does this already exist
Is there a bug in OpenReview for #UAI2024? We submitted 3 papers, and none of the reviewers updated their responses and there are only reject or accept decisions—no meta-review.
@marikgoldstein @brandondamos or solved through amortized inference 😎
1/What does it mean for an LLM to “memorize” a doc? Exactly regurgitating a NYT article? Of course. Just training on NYT?Harder to say We take big strides in this discourse w/*Adversarial Compression* w/@A_v_i__S @zhilifeng @zacharylipton @zicokolter 🌐:locuslab.github.io/acr-memorizati…🧵
GenAI does not stop, so it is time for a new blog post. Since LLMs are everywhere, I decided to take a look at them, my curious readers. And a bonus: An implementation of a teenyGPT 🤖✨ 📄Post: jmtomczak.github.io/blog/20/20_llm… 🖥️Code: github.com/jmtomczak/intr…
@marikgoldstein @ylecun @brandondamos lol I like this one :-)
@ylecun and then specified as a layer @brandondamos
Everything can be formulated as an optimization problem.
once @ylecun told me (heavily paraphrased), it's not F=ma but \min (F-ma)^2. i didn't realize its importance, but it is perhaps the most enlightning perspective i've ever heard.
Our new work on generative models for chemical reactions: much faster inference with flow matching (OT path) training scheme, better leveraging our knowledge about the problem is a key to solve science problems! Check out the paper if you are interested!
New paper alert: React-OT: Optimal Transport for Generating Transition State in Chemical Reactions (arxiv.org/abs/2404.13430). React-OT formulates TS search as a transport problem, approaching chemical accuracy while taking only 0.5 seconds in inference on a single GPU. #compchem
I2SB (i2sb.github.io) + (nearly optiaml) flow matching yields 1000x speed-up compared to standard denoising diffusion for generating highly accurate transition states 🧑🔬⚗️🧪 Check out our new preprint👇 arxiv.org/abs/2404.13430 Very fun collab w/ @chenru_duan @YuanqiD!
New paper alert: React-OT: Optimal Transport for Generating Transition State in Chemical Reactions (arxiv.org/abs/2404.13430). React-OT formulates TS search as a transport problem, approaching chemical accuracy while taking only 0.5 seconds in inference on a single GPU. #compchem
New paper alert: React-OT: Optimal Transport for Generating Transition State in Chemical Reactions (arxiv.org/abs/2404.13430). React-OT formulates TS search as a transport problem, approaching chemical accuracy while taking only 0.5 seconds in inference on a single GPU. #compchem
350 miles in four days, including two centuries, and I'm in DC!
Phi-3 just released by Microsoft. Three small size models (3.8B, 7B and 14B) trained on highly filtered and synthetic data. They report impressive performance since the 3.8B model (trained on 3T tokens) has MMLU of 69% matching Llama3 8B, and the 7B Phi-3 model has 75% MMLU,…
Excellent video-tutorial on the curse of unrolling. There's no feeling like when others build and improve upon your work 🤗
Check out my latest video on the "Curse of Unrolling," a counter-intuitive phenomenon when you unroll differentiate ("piggyback AD") through an iterative algorithm: youtu.be/80w5wDxq26c Even if your primal converges exponentially linear, the Jacobian initially does not. 🧵🧵
We have a new preprint out - your language model is not a reward, it’s a Q function! 1. The likelihood of the preferred answer must go down - it’s a policy divergence 2. MCTS guided decoding on language is equivalent to likelihood search on DPO 3. DPO learns credit assignment