-
Tweets581
-
Followers5K
-
Following581
-
Likes4K
Another thorny safety challenge for LLMs. Like Sleeper Agents (x.com/jayelmnop/stat…), @cem__anil has found behavior that is stubbornly resistant to finetuning. Training on MSJ shifts the intercept, but not the slope, of the relationship b/t # of shots and attack efficacy.
Another thorny safety challenge for LLMs. Like Sleeper Agents (x.com/jayelmnop/stat…), @cem__anil has found behavior that is stubbornly resistant to finetuning. Training on MSJ shifts the intercept, but not the slope, of the relationship b/t # of shots and attack efficacy. https://t.co/PXH5qhJS4A
Language models today are trained to reason either 1) generally, imitating online reasoning data or 2) narrowly, self-teaching on their own solutions to specific tasks Can LMs teach themselves to reason generally?🌟Introducing Quiet-STaR, self-teaching via internal monologue!🧵
I too have gotten Claude 3 to vertically center a <div>
I too have gotten Claude 3 to vertically center a <div>
gpt4: gets most of mmlu correct claude: gets most of mmlu correct gemini: gets most of mmlu correct mmlu: gets most of mmlu correct
Claude 3 Opus is great at following multiple complex instructions. To test it, @ErikSchluntz and I had it take on @karpathy's challenge to transform his 2h13m tokenizer video into a blog post, in ONE prompt, and it just... did it Here are some details:
Today, we're announcing Claude 3, our next generation of AI models. The three state-of-the-art models—Claude 3 Opus, Claude 3 Sonnet, and Claude 3 Haiku—set new industry benchmarks across reasoning, math, coding, multilingual understanding, and vision.
Have you ever done a dense grid search over neural network hyperparameters? Like a *really dense* grid search? It looks like this (!!). Blueish colors correspond to hyperparameters for which training converges, redish colors to hyperparameters for which training diverges.
How can we check LLM outputs in domains where we are not experts? We find that non-expert humans answer questions better after reading debates between expert LLMs. Moreover, human judges are more accurate as experts get more persuasive. 📈 github.com/ucl-dark/llm_d…
Jim Fan @DrJimFan
229K Followers 3K Following @NVIDIA Sr. Research Manager & Lead of Embodied AI (GEAR Lab). Creating foundation models for Humanoid Robots & Gaming. @Stanford Ph.D. @OpenAI's first intern.
Jacob Andreas @jacobandreas
14K Followers 958 Following Teaching computers to read. Assoc. prof @MITEECS / @MIT_CSAIL (he/him). https://t.co/5kCnXHjtlY https://t.co/2A3qF5vdJw
(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K Following
Percy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | Pianist
Felix Hill @FelixHill84
9K Followers 777 Following Research Scientist, Deepmind I try to think hard about everything I tweet, esp on 90s football and 80s music None of my opinions are really someone else's
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
Sam Bowman @sleepinyourhat
35K Followers 3K Following AI alignment + LLMs at NYU & Anthropic. Views not employers'. No relation to @s8mb. I think you should join @givingwhatwecan.
Edward 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.
Christopher Potts @ChrisGPotts
11K Followers 620 Following Stanford Professor of Linguistics and, by courtesy, of Computer Science, and member of @stanfordnlp and @StanfordAILab. He/Him/His.
Laura Ruis @LauraRuis
3K Followers 638 Following Currently research intern @cohere, PhD supervised by @_rockt and @egrefen. Language and LLMs. Spent time at FAIR, Google, and NYU (@LakeBrenden). She/her.
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
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.
Kayo Yin @kayo_yin
8K Followers 556 Following PhD student @berkeley_ai @berkeleynlp working on interpretability and signed languages. Former @msftresearch @deepmind @carnegiemellon @polytechnique. 🇫🇷🇯🇵
Horace He @cHHillee
23K Followers 449 Following Working at the intersection of ML and Systems @ PyTorch "My learning style is Horace twitter threads" - @typedfemale
rishi @RishiBommasani
4K Followers 2K Following Stanford CS PhD @StanfordCRFM @StanfordNLP @StanfordAILab @StanfordHAI Advisers: @percyliang @jurafsky Previous: @CornellCIS @clairecardie #FoundationModels
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.
Tim Rocktäschel @_rockt
29K Followers 1K Following Open-Endedness Team Lead @GoogleDeepMind, Professor of AI @AI_UCL, PI @UCL_DARK, @ELLISforEurope Scholar. ex @MetaAI (FAIR), @CompSciOxford. Opinions my own.
Andrew Lampinen @AndrewLampinen
7K Followers 1K Following Interested in cognition and artificial intelligence. Research Scientist @DeepMind. Previously cognitive science @StanfordPsych. Tweets are mine.
Stanford NLP Group @stanfordnlp
144K Followers 179 Following Computational Linguists—Natural Language—Machine Learning @chrmanning @jurafsky @percyliang @ChrisGPotts @tatsu_hashimoto @MonicaSLam @Diyi_Yang @StanfordAILab
Aryan Pandey (Look fo.. @AryanPa66861306
1K Followers 3K Following Half Machine Learning Engineer || DevOps and Machine Learning || Open Source at OpenVINO
Aman Chandra @amanchandra333
60 Followers 171 Following Software Developer | Robotics Enthusiast | Potterhead
Gagan Jain @gaganjain1582
50 Followers 745 Following Predoc Researcher @GoogleDeepMind | IIT Bombay'22
Sarah Wooders @sarahwooders
367 Followers 214 Following PhD @ucbrise @Berkeley_EECS working on systems for ML. Previously @glisten_ai (@ycombinator W20), CS/Math @MIT
esp @EspToTheFuture
1K Followers 3K Following 🚀building free & open source projects https://t.co/FbjSHFNbZ1,@dspacegame • videos @futuroptimist •🪐space,🌿plants, 🤖genai,💻software,⚙️hardware • loml❤️@fairyarcade 🥰
Josh Bickett @josh_bickett
7K Followers 1K Following New dad | Engineer @hyperwriteai @othersideai | On the side - experimenting with VLMs playing games
Anastasios Nikolas An.. @ml_angelopoulos
3K Followers 784 Following @Berkeley_EECS Ph.D. with Mike Jordan/Jitendra Malik. Conformal prediction, distribution-free uncertainty quantification, vision/imaging. Former @stanford_ee.
Abdi M. @scaredmonad
1K Followers 4K Following PLs, µ-compilers, type systems, λ-abstractions, trivia, thoughts @ https://t.co/x0BSfAqRpm
Ted Moskovitz @ted_moskovitz
740 Followers 192 Following PhD student at @GatsbyUCL. Formerly: intern at @DeepMind, @UberAILabs, student at @ColumbiaCompSci, @PrincetonNeuro.
Anurag Mishra @anuragm75160136
112 Followers 801 Following Building Scalable AI Applications | Senior Data Scientist @ EY | CSE Btech @ NIT MN | Linkedin: https://t.co/pCmSV6FmOe
Rohan Paul @rohanpaul_ai
12K Followers 812 Following ML Engineer (e/acc) 📌 https://t.co/x0IIWfnOt8 🚀 https://t.co/QEO4CKRl1b Open LLMs is Happiness 💡 Ex Deutsche & HSBC. DM for collaboration.
Liangyu Chen @cliangyu_
524 Followers 1K Following
Haoyi Fu @Haoyi_Fu
10 Followers 60 Following
Pankaj Gupta @pankaj_ipynb
28 Followers 920 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 😉.
Tessa Long @zhixuan_long
0 Followers 165 Following
Justin Friesen @justin__friesen
11 Followers 39 Following Just trying to learn about businesses and talk about cool stuff
Agent Columbus @AgentColumbus
14 Followers 63 Following
Ivelina Petrova @ivelinapetrovaX
34 Followers 785 Following Industrial Management and development Master Degree and Architect in Architecture [email protected] [email protected] [email protected]
Harsh Desai @dreamerharsh
1 Followers 3K Following
Jonathan hind @import_hind
75 Followers 958 Following
Jerry Hellden @Yariv_hellden
767 Followers 7K Following Founder @ Andromedaerospace, Experimental Aerospace engineer, Theoretical physicist, Quantum cosmologist, ML engineer, Futurist, Scientist, Reverse engineer🇺🇸
ララどり d/age IS.. @presklux49
149 Followers 553 Following シンギュラリタリアン。老化を治療し、永遠の若さを手に入れることを目指しています。老化研究を促進するツールとして、人工知能も重視しています。私の夢は、超知能が管理する色々な箱庭世界で、悠久の時を過ごすことです。
Rishabh Jain @RishabhJain_r
16 Followers 153 Following Software Engineer || Independent Researcher, AI and ML || IIT RPR’23
Vikram Dutt @vd_
816 Followers 7K Following
Human Feedback Founda.. @HumanFeedbackIO
37 Followers 28 Following Human Feedback Foundation provides human input to the open source AI community by building and supporting human feedback projects.
Noelle Nahhas @NahhasNoell
38 Followers 5K Following
Frank McGroarty @frankmcgroarty
134 Followers 1K Following
AI for Thinking @AIforThinking
31 Followers 684 Following
Liu Xiaochen @lxc0422
121 Followers 1K Following #ArtificialIntelligence #MachineLearning #ComputerVision #3Dreconstruction #3Dmeasurement #ImageProcessing #Robotics #Programming #Packaging
Juan Goldblatt @JuanGoldbl
89 Followers 5K Following
lentzl @ilelentzl
133 Followers 2K Following
Edmar Miyake @emiyake
38 Followers 462 Following
Jim Fan @DrJimFan
229K Followers 3K Following @NVIDIA Sr. Research Manager & Lead of Embodied AI (GEAR Lab). Creating foundation models for Humanoid Robots & Gaming. @Stanford Ph.D. @OpenAI's first intern.
Jacob Andreas @jacobandreas
14K Followers 958 Following Teaching computers to read. Assoc. prof @MITEECS / @MIT_CSAIL (he/him). https://t.co/5kCnXHjtlY https://t.co/2A3qF5vdJw
(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K Following
Percy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | Pianist
Sasha Rush @srush_nlp
52K Followers 464 Following Professor, Programmer in NYC. Cornell Tech, Hugging Face 🤗 https://t.co/cZl0wTfqGz
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.
Felix Hill @FelixHill84
9K Followers 777 Following Research Scientist, Deepmind I try to think hard about everything I tweet, esp on 90s football and 80s music None of my opinions are really someone else's
Christopher Manning @chrmanning
126K Followers 115 Following Director, @StanfordAILab. Assoc. Director, @StanfordHAI. Founder, @stanfordnlp. Prof. CS & Linguistics, @Stanford. IP @aixventureshq. 🇦🇺 Do #NLProc & #AI. 👋
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
Sam Bowman @sleepinyourhat
35K Followers 3K Following AI alignment + LLMs at NYU & Anthropic. Views not employers'. No relation to @s8mb. I think you should join @givingwhatwecan.
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).
Edward 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.
Christopher Potts @ChrisGPotts
11K Followers 620 Following Stanford Professor of Linguistics and, by courtesy, of Computer Science, and member of @stanfordnlp and @StanfordAILab. He/Him/His.
Laura Ruis @LauraRuis
3K Followers 638 Following Currently research intern @cohere, PhD supervised by @_rockt and @egrefen. Language and LLMs. Spent time at FAIR, Google, and NYU (@LakeBrenden). She/her.
Eric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0p
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.
Tal Linzen @tallinzen
16K Followers 894 Following Professor @nyuling and @NYUDataScience, research scientist @GoogleAI
Yoav Artzi @yoavartzi
13K Followers 163 Following Research/prof @cs_cornell + @cornell_tech🚡 / https://t.co/9YnWry7yHs / https://t.co/3VmRSyYm2d / asso. faculty director @arxiv / building https://t.co/f9QkzO5kaC
Eric Hambro @erichammy
540 Followers 1K Following member of technical staff @AnthropicAI formerly FAIR @MetaAI @Bloomberg @UCL @Cambridge_Uni @recursecenter opinions, regrettably, mine
david rein @idavidrein
2K Followers 983 Following Sentio ergo sum. AI alignment research at NYU, early employee @cohere
Emmanuel Ameisen @mlpowered
7K Followers 211 Following Research Engineer @AnthropicAI Previously: Staff ML Engineer @stripe, Wrote BMLPA by @OReillyMedia, Head of AI at @InsightFellows, ML @Zipcar
Orowa Sikder @OrowaSikder
1K Followers 304 Following the future could be amazing. let’s get to work | Research @AnthropicAI, ex: PhD @UCLCS
Anton Bakhtin @ SF @anton_bakhtin
2K Followers 127 Following MTS at @AnthropicAI, Ex @MetaAI, Ex @Google Three logicians walk into a bar ...
Chenlin Meng @chenlin_meng
8K Followers 833 Following Co-founder & CTO @pika_labs | ex @StanfordAILab @Stanford
Aaron Begg @aaron_begg
2K Followers 1K Following Community at @AnthropicAI | Chat with Claude: https://t.co/7w2gEKteuC | Build with Claude: https://t.co/ktsbQNA9D2
PatronusAI @PatronusAI
991 Followers 308 Following Automated evaluation for LLMs 🦄 Boost your confidence in generative AI ✨
Fei Xia @xf1280
6K Followers 694 Following Research Scientist at @GoogleDeepMind, Robot Learning, Computer Vision. PhD from @StanfordAILab @StanfordSVL, previously @Tsinghua_Uni. #AGI through Embodiment
Jascha Sohl-Dickstein @jaschasd
19K Followers 623 Following Member of the technical staff @ Anthropic. Most (in)famous for inventing diffusion models. AI + physics + neuroscience + dynamics.
John Thickstun @jwthickstun
1K Followers 535 Following Postdoc at Stanford. @StanfordCRFM @StanfordNLP @StanfordAILab Previous @uwcse @uw_wail Controllable Generative Models. AI for Music.
Joy He-Yueya @JoyHeYueya
72 Followers 68 Following CS PhD student working on AI for education @StanfordAILab
mrinank ⭐️ @MrinankSharma
816 Followers 436 Following alignment, poetry, soulmaking, devotion "live to the point of tears", camus
Dylan HadfieldMenell @dhadfieldmenell
2K Followers 2K Following Assistant Prof @MITEECS working on value (mis)alignment in AI systems; @[email protected] @[email protected] he/him
rohit @krishnanrohit
19K Followers 2K Following Building God at https://t.co/frWeoc7IVB - buy the book, it makes me happy! | essays weekly at https://t.co/TbCaC6VaaM
David Duvenaud @DavidDuvenaud
28K Followers 3K Following Machine learning prof @UofT. Working on generative models, inference, & latent structure.
jessica dai @jessicadai_
2K Followers 675 Following phd student @berkeley_ai !? also editorial @reboot_hq @kernel_magazine (she/her)
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)
Roger Grosse @RogerGrosse
10K Followers 750 Following
Gabe Grand @gabe_grand
949 Followers 281 Following Computation 🤖 & cognition 🧠 PhD student @MIT CSAIL
Katherine Lee @katherine1ee
6K Followers 932 Following understanding ourselves and our models. senior research scientist @GoogleBrain, @genlawcenter and @CornellCIS, formerly @Princeton @[email protected]
Prithviraj (Raj) Amma.. @rajammanabrolu
5K Followers 519 Following Interactive & grounded AI, RL, NLP. Assistant Prof @UCSanDiego. Research Scientist @DbrxMosaicAI. Prev: @allen_ai, @GeorgiaTech
Cem Anil @cem__anil
2K Followers 1K Following Machine learning / AI Safety at @AnthropicAI and University of Toronto / Vector Institute. Prev. student researcher @google (Blueshift Team) and @nvidia.
Atoosa Kasirzadeh @Dr_Atoosa
3K Followers 2K Following societal impacts of AI | asst Prof @EdinburghUni | research lead @CentreTMFutures, @turinginst | @GovAI_ fellow
Xindi Wu @cindy_x_wu
942 Followers 804 Following Data-centric multimodal ml PhD student @PrincetonCS, prev @RealityLabs @roboVisionCMU @CMU_Robotics @Snapchat
Sasha Sheng 🫶🏼 @hackgoofer
4K Followers 2K Following Builder, Dancer; @aiengfoundation & on a mission to help people be well. Lover of hackathons and updating my beliefs. Staying grounded. Prev: @MetaAI
Alireza Makhzani @AliMakhzani
2K Followers 894 Following Faculty Member @VectorInst, Associate Professor (status-only) @eceuoft, Canada CIFAR AI Chair
Frieda Rong @frieda_rong
332 Followers 959 Following CS PhD @Stanford, formerly 🚗 @UberATG, 🎓@UWaterloo.
Tristan Hume @trishume
6K Followers 330 Following Performance optimization lead @AnthropicAI. Profiling, distributed systems, dev tools, interpretability. [email protected]
jack morris @jxmnop
10K Followers 761 Following getting my phd in nlp @cornell_tech 🚠 // academic optimist // tweeting from the snack aisle at trader joes
Preetum Nakkiran @PreetumNakkiran
10K Followers 2K Following ML research @Apple. @sh_reya’s fiancé | PhD @Harvard, postdoc @UCSanDiego, EECS @Berkeley_EECS, "AI" @OpenAI, @GoogleAI
Rafael Rafailov @rm_rafailov
3K Followers 637 Following Ph.D. Student at @StanfordAILab. I work on Foundation Models and Decision Making. Previously @GoogleDeepMind @UCBerkeley
Amanda Askell @AmandaAskell
26K Followers 653 Following Philosopher & ethicist teaching models to be good @AnthropicAI. Personal account. All opinions come from my training data.
shreya rajpal @ShreyaR
6K Followers 767 Following ML, systems, and everything in between. Building @guardrails_ai. Previously founding eng @predibase, @Apple SPG, @driveai_, @IllinoisCS, @iitdelhi.
𝔊𝔴𝔢𝔯𝔫 @gwern
42K Followers 88 Following Internet besserwisser; pedantic, mean reply guy. 𝘞𝘢𝘵𝘢𝘴𝘩𝘪 𝘬𝘪𝘯𝘪𝘯𝘢𝘳𝘪𝘮𝘢𝘴𝘶! (Follow requests ignored due to terrible UI.)
Those of you who think AI can produce no stroke of genius, what human, pray, in the last 350 years of this portrait's existence conceived of such a refreshing elaboration?
We have finally done it. After all this time and due to countless requests from our users, we've shipped what I think is our most important and revolutionary feature yet. You can now interrupt Claude's yapping with our new stop generation button!
@alexalbert__ I hope @AnthropicAI realizes how much value you are contributing by making these updates relatable and being the voice for the community. keeps anthropic at top of mind a lot more between model updates.
Some personal updates: I joined OpenAI a few months ago, working on all things robustness/safety/privacy. Also, we are working to publish more of our safety work. See my first project here below, where we make initial progress on prompt injections and other attacks!
Introducing the Instruction Hierarchy, our latest safety research to advance robustness for prompt injections and other ways of tricking LLMs into executing unsafe actions. More details: arxiv.org/abs/2404.13208
✨🎓 I defended my dissertation “The Relationship between Linguistic Representations in Biological and Artificial Neural Networks” on Tuesday! 🎓✨ Incredibly grateful for my amazing PhD advisor @ev_fedorenko and a wonderful journey at @mitbrainandcog! 🧠🤖
In absolute awe at these old MIT course posters designed by Dietmar Winkler in the 60's:
@mattshumer_ Hey Matt, appreciate you bringing this to our attention. We haven't modified any of the Claude 3 models since we launched them. On claude.ai, there's currently two layers that may contribute to perceived model performance: our T&S measures (standard mechanisms…
Not everyone who dies gets to come back and tell their story, but thankfully Freddy did. A reminder to hold the people you love a little closer tonight, that medicine and health are the greatest gifts, and that at the end of the day, we're all patients. jamanetwork.com/journals/jaman…
Another gem to remember! The Mondrian Process papers.nips.cc/paper_files/pa…
New work on the Battleship Game accepted to CogSci '24! ⚓️🧠 How do people pose informative, grounded questions in uncertain environments? And how can we build machines that ask human-like questions? arxiv.org/abs/2402.19471 🧵 (1 / n)
new interview with @EthanJPerez (Anthropic) on the right attitude for alignment research: "this seems kind of plausible, I can do this finetuning run in one hour if I sit down and do it in a colab"
When I first saw Tree of Thoughts, I asked myself: If language models can reason better by searching, why don't they do it themselves during Chain of Thought? Some possible answers (and a new paper): 🧵
New Anthropic research: Measuring Model Persuasiveness We developed a way to test how persuasive language models (LMs) are, and analyzed how persuasiveness scales across different versions of Claude. Read our blog post here: anthropic.com/news/measuring…
They're making me tip my professor wtf @Stanford 😭
The next chapter about transformers is up on YouTube, digging into the attention mechanism: youtu.be/eMlx5fFNoYc The model works with vectors representing tokens (think words), and this is the mechanism that allows those vectors to take in meaning from context.
I like to think of myself as a researcher, but almost certainly the most valuable use of my time is writing US Visa letters.
few realize grimes invented accelerationism with her 2018 hit “we appreciate power”
i mean i called it "machine learning" until it started talking to me and then i thought it was fair to say "ai"
The rebranding of linear algebra as "artificial intelligence" may be the most successful marketing campaign of all time.
Made a short video exploring tool use and subagents! (w/ @aaron_begg and @typochondriac) Goal: Find the “quickest quicksort” implementation on GitHub by having a larger model orchestrate 100 subagent models Here’s how it works: 1/ x.com/anthropicai/st…
Tool use is now available in beta to all customers in the Anthropic Messages API, enabling Claude to interact with external tools using structured outputs.
Trends for United States
You might like
