Stephanie Chan @scychan_brains
Senior Research Scientist at DeepMind. Artificial and biological brains 🤖 🧠 Views are my own San Francisco, CA Joined November 2018-
Tweets468
-
Followers3K
-
Following2K
-
Likes3K
Looking forward to this ICML workshop on the unfolding future of sequence modeling.. state space models, long context, and more!
Looking forward to this ICML workshop on the unfolding future of sequence modeling.. state space models, long context, and more!
Come to Rotterdam and chat about in-context learning with us!
Come to Rotterdam and chat about in-context learning with us!
I’m excited to share that I’m working on a new book about building applications with foundation models! AI Engineering builds upon Machine Learning Systems Design, but with a focus on large scale, ready made models. The book covers: - The new AI stack (e.g. how it differs from…
Our new paper on AI persuasion, exploring definitions, harms and mechanisms. Happy to have contributed towards the section on mitigations to avoid harmful persuasion. Some highlights in 🧵 storage.googleapis.com/deepmind-media…
New Anthropic research: we find that probing, a simple interpretability technique, can detect when backdoored "sleeper agent" models are about to behave dangerously, after they pretend to be safe in training. Check out our first alignment blog post here: anthropic.com/research/probe…
🤯📽️🏃Annoucning controllable video generation workshop at ICML2024! Super excited for speaker line up, and looking forward to seeing submissions exploring controllability in video generation. Submission deadline is 31st May AOE (but why wait till then if you can submit now?!)
🤯📽️🏃Annoucning controllable video generation workshop at ICML2024! Super excited for speaker line up, and looking forward to seeing submissions exploring controllability in video generation. Submission deadline is 31st May AOE (but why wait till then if you can submit now?!)
More examples of our shoelace tying policy (4x speed):
Happy to host @ashrewards of @GoogleDeepMind at @raais 2024! Ashley works on reinforcement learning and foundational world models. She was part of the team that created Gato, a multi-modal, multi-task, multi-embodiment agent that was able to perform a diverse range of tasks,…
Thrilled to share a review on THE LANGUAGE NETWORK AS A NATURAL KIND—a culmination of ~20 yrs of thinking about+studying language from linguistic, psycholinguistic, and cog neuro perspectives. @NatRevNeurosci rdcu.be/dEylV With the amazing @neuranna @tamaregev 🥳 🧵1/n
Our new paper delves into the circuits and training dynamics of transformer in-context learning (ICL) 🥳 Key highlights include 1️⃣ A new opensourced JAX toolkit that enables causal manipulations throughout training 2️⃣ The toolkit allowed us to "clamp" different subcircuits to…
Our new paper delves into the circuits and training dynamics of transformer in-context learning (ICL) 🥳 Key highlights include 1️⃣ A new opensourced JAX toolkit that enables causal manipulations throughout training 2️⃣ The toolkit allowed us to "clamp" different subcircuits to…
Standard transformer-based language models use the same amount of compute for each token. Our new method, which we call Mixture-of-Depths, allows transformers to instead learn to dynamically allocate compute to specific positions in a sequence. arxiv.org/abs/2404.02258
Our new AI model Genie can create playable worlds in the style of 2D platformers - all from a single image prompt, sketch or text description. ✏️ As a foundation world model, Genie could also help us train AI agents. Here's how. ↓ technologyreview.com/2024/02/29/108…
Introducing Yell At Your Robot (YAY Robot!) 🗣️- a fun collaboration b/w @Stanford and @UCBerkeley 🤖 We enable robots to improve on-the-fly from language corrections: robots rapidly adapt in real-time and continuously improve from human verbal feedback. YAY Robot enables…
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'sEric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0pKyunghyun 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).Andrew Lampinen @AndrewLampinen
7K Followers 1K Following Interested in cognition and artificial intelligence. Research Scientist @DeepMind. Previously cognitive science @StanfordPsych. Tweets are mine.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.Behnam Neyshabur @bneyshabur
18K Followers 690 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingJeff Dean (@🏡) @JeffDean
296K Followers 6K Following Chief Scientist, Google DeepMind and Google Research. Co-designer/implementor of things like @TensorFlow, MapReduce, Bigtable, Spanner, Gemini .. (he/him)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/2A3qF5vdJwEugene 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.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.Hattie Zhou @oh_that_hat
5K Followers 765 Following Finding \hat{y} Give me anonymous feedback: https://t.co/7aBNrpbad8Yi Tay @YiTayML
29K Followers 97 Following chief scientist / cofounder @RekaAILabs 🫠 past: research scientist @google brain 🤯 currently learning to be a dad 🍼Tal Linzen @tallinzen
16K Followers 893 Following Professor @nyuling and @NYUDataScience, research scientist @GoogleAITim Dettmers @Tim_Dettmers
29K Followers 821 Following PhD Student at @UW. I blog about deep learning and PhD life at https://t.co/Y78KDJJFE7.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.Dileep George @dileeplearning
10K Followers 1K Following AGI research @DeepMind. Ex cofounder & CTO @vicariousai (acqd by Alphabet) and @Numenta. Triply EE (BTech IIT-Mumbai, MS&PhD Stanford). #AGIComicsKeerthana Gopalakrish.. @keerthanpg
13K Followers 824 Following Building Embodied AGI. Research @DeepMind. Opinions my own.Naomi Saphra @nsaphra
7K Followers 1K Following Waiting on a robot body. ML/NLP. All opinions are universal and held by both employers and family. Same username on every lifeboat off this sinking ship.Jack Rae @drjwrae
9K Followers 354 Following Principal Scientist @ Google DeepMind Work on Gemini 💎♊ Compression is all you need LLMs (e.g. Gopher, Chinchilla, Gemini) 💼 Past: OpenAI, QuoraKunvar Thaman @firstuserhere
220 Followers 640 Following Taking apart neural networks and putting them back together for a living Social profiles: https://t.co/OxoeMvCw3aStefan Juang @StefanJuang
145 Followers 1K Following The final goal of AI is not just to create intelligent machines, but to understand intelligence itself.Suyog Chandramouli @suyoghc
518 Followers 5K Following Cognition, interaction, and statistics. Postdoctoral researcher @FCAI_fi @AaltoUniversity PhD @IUBloomingtonRoger Beaty @Roger_Beaty
3K Followers 2K Following Director of the Cognitive Neuroscience of Creativity Lab and Assistant Professor of Psychology at Penn State UniversityXuhui Zhang @XuhuiZhangXHZ
7 Followers 252 FollowingGuillaume Desjardins @gdesjardins_ml
181 Followers 208 FollowingDana Mahmood @deordered
22 Followers 720 Following Fine-tuning AI models oftentimes & practicing philosopher at other times.ELON MUSK🌐 @elonreeevvemusk
2 Followers 167 Following CO-FOUNDER OF NEURALINK AND OPENAI; AND PRESIDENT OF THE MUSK FOUNDATION. THE CEO TECHNOLOGY OFFICERS OF SPACEX AND TESLA.🌏🚀🚀Olusegun Ode @OlusegunOde2
241 Followers 1K FollowingSheteh @Sheteh457480
0 Followers 140 FollowingAyoub Elmendoub @ElmendoubA
3 Followers 112 FollowingElonmusk @Elonmusk220790
7 Followers 731 Following ELON MUSK🌐 CEO - Twitter, SpaceX🚀, Tesla🚘 Founder - The Boring Company🛣️ Co-founder - Neuralink, OpenAl🤖 @teslamotors @elon.musk_oficialDamaris Wright @DamarisWri63196
123 Followers 3K FollowingAvery Ryoo @averyryoo
117 Followers 518 Following MSc @Mila_Quebec + @UMontrealDIRO | NeuroAI, representation learning, cogsci, and Toronto sports teamsTiwa Eisape @tiwa_eisape
1K Followers 1K Following PhD student at @MIT working on NLP and cognitive science - @NSF grfp fellow. Previously with @GoogleAI and @Meta FAIRABHISHEK KUMAR @abhishekkr8399
41 Followers 3K Following Competitive Programmer | Software Engineer (Fresher) | Strong Analytical & Problem-Solving Skills | Web & Mobile Development ExperienceSeeTweet @stweeeeeet
29 Followers 285 FollowingKevin Pham @Hsiejvdos
0 Followers 12 FollowingJason Phang @zhansheng
3K Followers 1K Following Policy Research at @OpenAI. PhD @NYUDataScience, @AiEleuther, 🇸🇬. Prev: @Google, @MicrosoftKanishk Gandhi @gandhikanishk
921 Followers 692 Following Phd @Stanford CS; w/ Noah Goodman, Dorsa Sadigh | Prev: @LakeBrenden @NYUDataScience, @IITKanpur, @Path_AIErik Scarlatescu @saerik2001
4 Followers 38 FollowingHoward Chen @__howardchen
855 Followers 1K Following PhDing @princeton_nlp & @PrincetonPLI. Previously: Meta AI (intern) / ASAPP research / Cornell Tech / NTU (Taiwan).Abdulrahman Tabaza @embed_dim
4 Followers 799 Following enjoyer of various vector spaces, encoders and modalitiesDr. Cristina Vanbergh.. @VanberghenEU
42K Followers 34K Following Professor - now EUI Florence, Senior Expert @EU_Commission, diplomat, Stanford Center for Internet, @StanfordCIS EUinfluencer 2019-23 @ZNConsulting, PhD LeidenSatish Venkatesan @satishKvenk
54 Followers 1K Followingzhou su @suhmily
31 Followers 358 FollowingMartin Fan @perfectoid_ai
394 Followers 8K FollowingErik Nijkamp @erik_nijkamp
1K Followers 765 Following Ph.D., Generative Modeling, Representation Learning. Lead Research Scientist at Salesforce Research. Creator of CodeGen. Co-Creator of ProGen2, XGen LLMs.hypervanse @hypervanse
91 Followers 653 Following PhD in Physics, researcher, programmer, musical artist, gamer... 超人I07XNbUI4 @DeepFeed2
48 Followers 3K FollowingArpit Bansal @arpitbansal297
1K Followers 835 Following PhD Candidate @UMDCS. Past @AmazonScience, @IITKgp. A brick in the creation of Artificial General Intelligence.larry_yang @larry_x_yang
0 Followers 46 FollowingTimothy Nguyen @IAmTimNguyen
7K Followers 414 Following Machine learning researcher at @GoogleDeepMind, mathematician, quantum physicist. Host of The Cartesian Cafe podcast. All opinions are my own.taesiri @taesiri
529 Followers 4K Following PhD Student at UofA, Working on Large Multimodal Models.wanlin zhu @neuromanifold
32 Followers 3K FollowingAsma Ghandeharioun @ghandeharioun
2K Followers 489 Following Research Scientist @GoogleAI working on ML interpretability & human-centered AI, PhD from @MITAlpay Ariyak @AlpayAriyak
1K Followers 2K Following AI @RunPod_io | Lead: @OpenChatDev (600k+ downloads on HuggingFace🤗)sritee @Sridhaar96
99 Followers 627 Following Interested in ML and Robotics. Researcher Engineer @DeepMind.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'sEric Jang @ericjang11
69K Followers 3K Following physical AGI at 1X. Author of "AI is Good for You" https://t.co/eFg4WXhg0pRosanne Liu @savvyRL
33K Followers 968 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(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K FollowingKyunghyun 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).Sergey Levine @svlevine
80K Followers 122 Following Associate Professor at UC Berkeley Co-founder, Physical IntelligenceShane Gu @shaneguML
28K Followers 1K Following Research Scientist & Manager @GoogleDeepMind Tokyo/MTV. ex: @GoogleAI Brain, @OpenAI. (JP: @shanegJP)Anthropic @AnthropicAI
262K Followers 26 Following We're an AI safety and research company that builds reliable, interpretable, and steerable AI systems. Talk to our AI assistant Claude at https://t.co/aRbQ97uk4d.Percy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistAndrew Lampinen @AndrewLampinen
7K Followers 1K Following Interested in cognition and artificial intelligence. Research Scientist @DeepMind. Previously cognitive science @StanfordPsych. Tweets are mine.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.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 🌈Soumith Chintala @soumithchintala
186K Followers 883 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 446 Following Researcher (Google DeepMind/Brain in Zürich, ex-RWTH Aachen), Gamer, Hacker, Belgian. Mostly gave up trying mastodon as [email protected]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. 🇵🇱🇺🇸Danijar Hafner @danijarh
14K Followers 869 Following Building AI that makes autonomous decisions using world models, artificial curiosity, and temporal abstraction @DeepMindDavid 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 🦋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.Roger Beaty @Roger_Beaty
3K Followers 2K Following Director of the Cognitive Neuroscience of Creativity Lab and Assistant Professor of Psychology at Penn State UniversityJason Phang @zhansheng
3K Followers 1K Following Policy Research at @OpenAI. PhD @NYUDataScience, @AiEleuther, 🇸🇬. Prev: @Google, @MicrosoftBrian Ichter @brian_ichter
2K Followers 178 Following Research Scientist at Google Brain, interested in robotics and AIDaniel Johnson @_ddjohnson
2K Followers 576 Following Researcher at @GoogleDeepMind. PhD student at @VectorInst / @UofT. Building tools to study neural nets and find out what they know. He/him.Yevgen Chebotar @YevgenChebotar
1K Followers 286 Following Robot AI @Figure_robot | Former Research Scientist @GoogleDeepmind | 🤖 🦾Arpit Bansal @arpitbansal297
1K Followers 835 Following PhD Candidate @UMDCS. Past @AmazonScience, @IITKgp. A brick in the creation of Artificial General Intelligence.sritee @Sridhaar96
99 Followers 627 Following Interested in ML and Robotics. Researcher Engineer @DeepMind.Duane Watson @duane_g_watson
5K Followers 2K Following Psycholinguist, professor, Vanderbilt (he, him, his)Resemble AI @resembleai
4K Followers 24 Following High-quality AI voice generator that captures human emotion- as seen in #TheAndyWarholDiaries on Netflix. https://t.co/80Js2ahx4tRunway @runwayml
185K Followers 300 Following An applied AI research company building for the next era of art, entertainment and human creativity. We're hiring: https://t.co/Aj11xyhxOgNikolaus West @NikolausWest
722 Followers 422 Following Multimodal seeing tools for robot and AI builders https://t.co/f4IweMLYwrRajko Radovanović @rajko_rad
4K Followers 4K Following AI/infra @a16z (partner to amazing teams eg @MistralAI @udiomusic); Enjoy most things outdoors, care about democracy in 🇷🇸🇭🇷🇸🇮🇧🇦🇲🇪Andrew Sanchez @avincentsanchez
310 Followers 142 Following COO and Co-Founder at @udiomusic | Oxford DPhil | HarvardYaroslav Ganin @yaroslav_ganin
4K Followers 231 Following Co-Founder @udiomusic. Research Scientist. Previously: @DeepMindAI, Mila (Montréal, Canada), Skoltech (Moscow, Russia). Views are my own.udio @udiomusic
28K Followers 0 FollowingPeter J. Liu @peterjliu
4K Followers 2K Following Research Scientist @ Google B̵r̵a̵i̵n̵ DeepMind, frontier language models research (aka chatbot engineer). Opinions are my own. 🤖🔄🚀Kanishk Gandhi @gandhikanishk
921 Followers 692 Following Phd @Stanford CS; w/ Noah Goodman, Dorsa Sadigh | Prev: @LakeBrenden @NYUDataScience, @IITKanpur, @Path_AISholto Douglas @_sholtodouglas
15K Followers 858 Following Scaling Gemini @Deepmind - working towards intelligence too cheap to meterMohamed Elhoseiny @moElhoseiny
1K Followers 761 Following AI Prof @KAUST_news supporting @KAUSTVisionCAIR lab (hiring!), AI artist; formerly @StanfordGSB Igniter,@facebookai,@Baidu,@Adobe,@RutgersUsam ritter @ritterstorm
267 Followers 181 Following Currently into episodic memory for deep RL agents, planning, and meta-RL. On the neuroscience team at DeepMind.Peter Humphreys @p_humphreys
55 Followers 34 Following AI, quantum and neuroscience. Scientist @ DeepMindยุ้ย Yada Pru.. @yadapruksachatk
641 Followers 420 Following nlp scientist by passion, builder at heart. Prev @AlexaAI @CILVRatNYU @kp_fellows 🇹🇭 ☀️🏡Adam Karvonen @a_karvonen
1K Followers 299 Following Interested in ML and software. I prefer email to DM.The Information @theinformation
96K Followers 697 Following The leading publication high-powered tech executives and founders read daily.Joe Edelman @edelwax
8K Followers 1K Following wise AI; moral graphs; mechanism & game design; big data virtue ethics; meaning metrics; values-based choice theory @meaningalignedMeaning Alignment Ins.. @meaningaligned
934 Followers 10 Following The Meaning Alignment Institute is a research organization with the goal of ensuring human flourishing in the age of AGI.Dan Roberts @danintheory
4K Followers 572 Following I studied gravity. AI fellow @sequoia + researcher @mit physics. Co-founded @diffeo, acquired by @salesforce. Co-author "The Principles of Deep Learning Theory”Mandiant @Mandiant
125K Followers 4K Followingj⧉nus @repligate
16K Followers 1K Following ⌥ Breach Mystic ⌥ Heisenbergian Harlequin ⌥ Schrodingerian Godflipper ⌥ Rabbit-Hole-As-A-Service (RHAAS)80,000 Hours @80000Hours
29K Followers 536 Following You have 80,000 hours in your career. This makes it your best opportunity to have a positive impact on the world.Open Philanthropy @open_phil
15K Followers 17 Following Open Philanthropy's mission is to help others as much as we can with the resources available to us.Effective Altruism @EffectvAltruism
23K Followers 98 Following Effective altruism is a community that uses evidence and reason to find the best ways to improve the world. Retweet ≠ endorsement. (Account run by @CentreforEA)Toby Ord @tobyordoxford
17K Followers 138 Following Senior Researcher at Oxford University. Author — The Precipice: Existential Risk and the Future of Humanity.Garry Tan @garrytan
433K Followers 4K Following President & CEO @ycombinator —Founder @Initialized—PM/designer/engineer who helps founders—YouTuber—San Francisco Democrat accelerating the boom loop—e/accKerry Wang @kerryxwang
375 Followers 106 Following Co-founder, CEO of @searchlight_ai Backed by @Accel, @FoundersFund, & @YCombinatorAnna Wang @AnnaXWang
265 Followers 129 Following Head of AI @JoinMultiverse | fmr. co-founder & CTO @searchlight_ai (acquired) backed by @Accel, @FoundersFund, & @ycombinatorDwarkesh Patel @dwarkesh_sp
55K Followers 700 Following Being pretrained Host of Dwarkesh Podcast https://t.co/3SXlu7fy6N https://t.co/rEhnfYywXY https://t.co/hQfIWdM1UnKatja Grace 🔍 @KatjaGrace
8K Followers 798 Following Thinking about whether AI will destroy the world at https://t.co/pMilDvd4ya. DM or email for media requests. Feedback: https://t.co/zGAm1i7SKHKangwook Lee @Kangwook_Lee
2K Followers 675 Following Assistant Professor, ECE, UW-Madison / Leading deep learning research @ KRAFTONDelighted to share ✨Med-Gemini✨ - our new family of multimodal models for medicine unlocking new possibilities for health - arxiv.org/pdf/2404.18416 More accurate multimodal conversations about medical images🩻, surgical videos📽️, genomics🧬, ultra-long health records📚, ECGs🫀…
We are thrilled to announce that our Director, Dr. Joshua Langberg, has been promoted to Chief Wellness Officer (CWO) of @RutgersU-New Brunswick! 🎉 He will play a pivotal role in the new #ScarletWell initiative promoting behavioral & mental health across campus. Congratulations!
News & Views from Justin Wood, framing Emin's article perfectly: "To date, the nature-nurture debate largely stems from different intuitions about the nature of the experiences available for learning." Now, we can test these intuitions computationally! nature.com/articles/s4225…
with speakers: @scychan_brains @996roma @jcrwhittington @GretaTuckute @RTomMcCoy and panelists: @morganbarense @m_heilb @cocosci_lab @LakeBrenden
3. When and why do natural and artificial systems rely on in-context versus in-weights learning? 4. How does in-context learning relate to other concepts from cognitive science? and many others!
We have gathered speakers from computer science, linguistics, psychology, and neuroscience to map out questions such as: 1. How well can human learning be modeled using in-context learning? 2. Which neural architectures support in-context learning?
Excited to announce our full-day workshop on “In-context learning in natural and artificial intelligence” at CogSci (@cogsci_soc) 2024 in Rotterdam (with @JacquesPesnot @akjagadish @summerfieldlab and Ishita Dasgupta). jacquespesnot.github.io/2024_CogSci_Wo…
I am honored to announce that we will organize the "Next Generation of Sequence Modeling Architectures" workshop at ICML 2024 this year in Vienna. The workshop will take place on Friday, the 26th of July. The workshop page is: sites.google.com/view/ngsmworks…
We will have an amazing line of speakers, including @_albertgu, @scychan_brains, @sohamde_, @HochreiterSepp, and many more exciting speakers! Covering a range of topics on developing the next generation of sequence models has been a topic near my heart for more than 10 years🙂.
@scychan_brains super interesting - and has implications for safety, and when/ where best to "bake it in" to large models!
Hard to believe it needs to be said, but universities should not call the police to arrest protesting students. Very disappointing this happened at NYU, and alarming that one of the justifications for asking the police to step in was the students' "intimidating chants".
There is a really nice community of researchers developing transformer alternatives. Want to highlight these impressive folks. Simran Arora (@simran_s_arora), Chunting Zhou (@violet_zct), Dan Fu (@realDanFu), and Songlin Yang (@SonglinYang4)
I’m excited to share that I’m working on a new book about building applications with foundation models! AI Engineering builds upon Machine Learning Systems Design, but with a focus on large scale, ready made models. The book covers: - The new AI stack (e.g. how it differs from…
So who at @AnthropicAI received this before me? 🤔
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
SoTA LLMs typically exhibit 99%+ non-zero activations, but it turns out that they are still intrinsically quite sparse! We introduce CATS, a simple post-training technique that achieves 50% activation sparsity for MLP layers with almost no drop in downstream evals, while…
Our new paper on AI persuasion, exploring definitions, harms and mechanisms. Happy to have contributed towards the section on mitigations to avoid harmful persuasion. Some highlights in 🧵 storage.googleapis.com/deepmind-media…
@bmorphism @avisingh599 @l32zhang @scychan_brains @ankesh_anand @zaheersm_1 @Azade_na @its_ericchu @FeryalMP @AleksandraFaust @hugo_larochelle Only one way to find out ;)
New Anthropic research: we find that probing, a simple interpretability technique, can detect when backdoored "sleeper agent" models are about to behave dangerously, after they pretend to be safe in training. Check out our first alignment blog post here: anthropic.com/research/probe…
LLMs process text from multiple sources and may face conflicting instructions. We teach our models to follow instructions from the highest priority input, giving better defense against attacks. Our Safety Systems team is hiring: openai.com/careers/search…