Niklas Stoehr @niklas_stoehr
PhD student @ETH advised by @RyanDCotterell, @Cervisiarius and @AaronSchein ⭕️ Language Model Interpretability and Application ⭕️ @GoogleAI, @TechAtBloomberg niklas-stoehr.com Zurich, Switzerland Joined October 2017-
Tweets236
-
Followers789
-
Following744
-
Likes4K
How much does an LM depend on information provided in-context vs its prior knowledge? Check out how @vesteinns, @niklas_stoehr, @JenniferCWhite, @AaronSchein, @ryandcotterell + I answer this by measuring a *context's persuasiveness* and an *entity's susceptibility*🧵
Excited to finally have @miserlis_ visiting us at @ETH_AI_Center and ETH D-GESS sharing his cool work on „Social Science as a Problem Space for NLP“, co-hosted by @ryandcotterell and @ellliottt.
If you are at #EACL2024, on Friday 4pm Malta time, I will give a (virtual) talk on 𝗖𝗼𝗻𝘁𝗿𝗼𝗹𝗹𝗲𝗱 𝗗𝗲𝗰𝗼𝗱𝗶𝗻𝗴 𝗳𝗿𝗼𝗺 𝗟𝗮𝗻𝗴𝘂𝗮𝗴𝗲 𝗠𝗼𝗱𝗲𝗹𝘀 at the 1𝘴𝘵 𝘗𝘦𝘳𝘴𝘰𝘯𝘢𝘭𝘪𝘻𝘢𝘵𝘪𝘰𝘯 𝘰𝘧 𝘎𝘦𝘯𝘦𝘳𝘢𝘵𝘪𝘷𝘦 𝘈𝘐 (𝘗𝘌𝘙𝘚𝘖𝘕𝘈𝘓𝘐𝘡𝘌) 𝘸𝘰𝘳𝘬𝘴𝘩𝘰𝘱.
In today's Oral Presentation session at #EACL2024 (11:30 CET), @ETH_en's @niklas_stoehr will present "Unsupervised Contrast-Consistent Ranking with Language Models," a paper co-authored w/ our #AI researchers Pengxiang Cheng, Jing Wang, @daniel_preotiuc & Rajarshi Bhowmik #NLProc
Congratulations to @niklas_stoehr of @ETH_en who co-authored #EACL2024 paper "Unsupervised Contrast-Consistent Ranking with Language Models" together with our #AI researchers Pengxiang Cheng, Jing Wang, @daniel_preotiuc & Rajarshi Bhowmik bloom.bg/3VnM0DA #NLProc (1/2)
🏛📣 Life Update 📣🏛 This fall, I’ll be starting a position as an Assistant Professor of Computational Linguistics at @Georgetown University! I’m excited to be moving to DC🇺🇸 and to be joining many wonderful colleagues at @GU_Linguistics
Happy to share the good news that as of today I’m officially tenured — aka promoted from ass. prof. to ass. prof. 🤔 ethrat.ch/en/appointment… Many thanks to all those who have been supporting me along this exciting path: students, colleagues, and most of all my lovely family. 💋
My ex EPFL lab mates @krisgligoric and @tizianopiccardi invited me to lead a session on “Making Measurements with LLMs” in their Social NLP reading group tomorrow: sites.google.com/view/social-nlp 😍 I’ll present arxiv.org/pdf/2309.06991… Let me know if you are @Stanford and free to chat.
Preprint for this work (led by @seanohagann) is now online: arxiv.org/pdf/2312.09203… Feedback very welcome!
Preprint for this work (led by @seanohagann) is now online: arxiv.org/pdf/2312.09203… Feedback very welcome!
🚨New round of recruiting! Join our growing team to bridge ML models with interpretable domain expertise. Learn more: neuroexplicit.org Apply by Jan 7
🚨New round of recruiting! Join our growing team to bridge ML models with interpretable domain expertise. Learn more: neuroexplicit.org Apply by Jan 7
Thank you to #EMNLP2023 chairs for the 😱 two 😱 outstanding paper awards! I am so grateful to have worked on these projects with wonderful colleagues — @tpimentelms (who is the first author on one of the papers!), @clara__meister, @kmahowald and @ryandcotterell
… just so sad to miss this amazing program at the @BlackboxNLP workshop at #EMNLP2023! Invited talks by @ZhijingJin and @ABosselut (hop 🇨🇭!) and a panel discussion on Mechanistic Interpretability involving @NeelNanda5, @johnhewtt and @boknilev. ◼️🔍
… just so sad to miss this amazing program at the @BlackboxNLP workshop at #EMNLP2023! Invited talks by @ZhijingJin and @ABosselut (hop 🇨🇭!) and a panel discussion on Mechanistic Interpretability involving @NeelNanda5, @johnhewtt and @boknilev. ◼️🔍
Our #AI Engineering Group' s Sr. Research Scientist @daniel_preotiuc will give an Industry Keynote on "Modular Language Modeling through Model Merging" at today's 2023 Singapore Symposium on Natural Language Processing (16:30 SGT) bloom.bg/47INeN1 #nlproc #EMNLP2023 #SSNLP
Ryan David Cotterell @ryandcotterell
9K Followers 1K FollowingJosef Valvoda @ValvodaJosef
674 Followers 1K Following PhD candidate @CambridgeNLP group @Cambridge_UniMrinmaya Sachan @mrinmayasachan
2K Followers 2K Following Assistant Professor of Computer Science at ETH Zurich working in natural language processing (#NLProc), machine learning and education (#edtech).Kumar Shridhar @JupyterAI
584 Followers 1K Following PhD in ML/NLP @eth_en | I do #NLProc and #AI | Past: @MSFTResearch @AIatMeta @AmazonScience @rptu_kl_ld | He/him. Views are my own.Tiago Pimentel @tpimentelms
1K Followers 248 Following Postdoc at @ETH_en. Formerly, PhD student at @Cambridge_Uni.Manoel @manoelribeiro
3K Followers 1K Following CS PhD student @ EPFL — On the job market for 2023-2024! 🐘: @[email protected] Keywords: Computational Social Science, Platforms, Communities, ModerationSebastian Gehrmann @sebgehr
5K Followers 2K Following Head of NLP, CTO office, @Bloomberg. (he/him) Generating natural language, one word at a time. Also making sense of that language afterwards. views my ownClara Isabel Meister @clara__meister
2K Followers 49 Following PhD student in the ML Institute at ETH Zurich. Still figuring out how Twitter works... 🤦♀️Christine de Kock @christinedekock
258 Followers 254 Following Researcher & lecturer in NLP at @unimelbMachel Reid @machelreid
2K Followers 1K Following Research Scientist @GoogleDeepMind Working on LLMs on the Gemini Team; did gemini 1.5 proPatrick Haller @padraiglindrome
261 Followers 940 Following PhD Student in Computational Linguistics @cl_uzh. Interested in language modeling, human language processing... and drag race I guess. he/himPasquale Minervini �.. @PMinervini
7K Followers 4K Following Researcher in ML/NLP at the University of Edinburgh (faculty @InfAtEd @EdinburghNLP), @ELLISforEurope, @UCL_NLP, PI for @Clarify2020, https://t.co/WydvfU8ugz he/theyLucy Li @lucy3_li
4K Followers 2K Following @UCBerkeley PhD student + @allen_ai. Human-centered #NLProc, computational social science, AI fairness. she/her. https://t.co/rtSSUhWQnLSwabha Swayamdipta @swabhz
6K Followers 461 Following Assistant Prof. @CSatUSC | Researcher in #NLProc | Previously with @uwnlp @allenai | she/herGabriele Sarti @gsarti_
2K Followers 2K Following PhD Student @GroNLP 🐮, core dev of @InseqLib (https://t.co/tTjrg26ygQ). Interpretability ∩ HCI ∩ #NLProc. Prev: @AmazonScience, @Aindo_AI, @ItaliaNLP_Lab.Zhijing Jin @ZhijingJin
3K Followers 1K Following Final-year PhD @MPI_IS & @ETH_en w/ @bschoelkopf. Research on (1) @CausalNLP and (2) NLP4SocialGood @NLP4SG. Mentor and mentee @ACLMentorship.Javier Ferrando @javifer_96
277 Followers 480 Following PhD Student @la_UPC. Interpretability in NLPViviana @Viviana75842443
2 Followers 161 FollowingVésteinn Snæbjarnar.. @vesteinns
134 Followers 421 Following PhD fellow at @AiCentreDK @ELLISforEurope @DIKU_institut @BelongieLab • NLP & CVTim Davidson @im_td
537 Followers 396 Following PhD research @EPFL on reliable magic | machine learning & company building | @nyuniversity @UvA_Amsterdam alumn | mostly tweet about AI + some wild thoughtsClément Dumas @Butanium_
78 Followers 217 Following CS MSc student at ENS Paris-Saclay Ecosystem simulation enjoyer/Aspiring AI safety researcherMargaretJannett @JannettMar18405
5 Followers 732 FollowingEdoardo Debenedetti @edoardo_debe
749 Followers 2K Following CS PhD student @CSatETH 🇨🇭 | ML Security and Privacy 💻🕵️♂️ | prev @EPFL_en @PoliTOnews | Help 🇺🇦 on https://t.co/32YZoUP39z | From 🇪🇺🇮🇹Pensé FFun @inftyCategory
100 Followers 6K FollowingAdi Haviv @adihaviv
439 Followers 268 Following CS Ph.D. Candidate at @TelAvivUni. Researching #NLProc and Computer Vision.Veniamin Veselovsky @VminVsky
296 Followers 528 Following pre-trained on epfl and university of torontoVarvara Arzt @wienergespenst
63 Followers 314 Following NLP researcher at TU Wien & AAU, predoc — love languages, maths, programming, sailing, music, and scents ☮️ Views are mineHarshali Ranjan @harshaliranjan
0 Followers 83 FollowingYihuai Hong @YihuaiH91773
25 Followers 157 Following CS Undergraduate interested in NLP research @SCUT previously Research Intern in @UCLShayne Longpre @ShayneRedford
4K Followers 997 Following PhD @MIT. Prev: @Google Brain, @apple ML, @stanfordnlp. 🇨🇦 Interests: AI/ML/NLP, Data-centric AI, transparency & societal impactJordan Gong @jordan__gong
42 Followers 2K FollowingSerhan Yilmaz @srhnylmz14
67 Followers 771 Following current junior cs undergrad @sabanciu & president/founder @kaisabanci // prev @EPFL @YapiKredi @BU_Tweets @kocuniversity // contact: dmVishala Mishra @vishala_mishra
272 Followers 3K Following Physician and Clinical Informaticist interested in using data for health equity.Asma Ghandeharioun @ghandeharioun
2K Followers 489 Following Research Scientist @GoogleAI working on ML interpretability & human-centered AI, PhD from @MITJohn Zhang @JohnZha78551114
31 Followers 149 FollowingYangsibo Huang @YangsiboHuang
1K Followers 726 Following PhD candidate @Princeton. Prev: @GoogleAI @AIatMeta.Tanmay Parekh @tparekh97
450 Followers 398 Following PhD student at @UCLA | MLT @LTIatCMU | Applied Scientist @amazonIN | BTech @iitbombayTheLobbyistGuy @TheLobbyistGuy
679 Followers 2K Following Bringing AI to K Street with @LobbyMaticAIKrystof Mitka @krystof_mitka
113 Followers 512 Following Currently completing undergraduate double degree in Applied Mathematics and Computer Science in 🇳🇱Patrick Y. Wu @PatrickYWu
467 Followers 1K Following postdoc @CSMaP_NYU | computational social scientist working on AI+NLP | PhD/MA @UMich, BA @UChicago | https://t.co/yjLeBlS86oAmr Khalifa @AmrMAlameen
823 Followers 393 Following I do AI research and other cool stuff @Google, Google DeepMind team @DeepMind, also PhD student @Mila_Quebec | Opinions are my own |Francesco Ignazio Re @francignare
1 Followers 15 FollowingMarc Marone @ruyimarone
420 Followers 586 Following PhD student at Johns Hopkins @jhuclsp. Previously @microsoft Semantic Machines, @mstranslator, @GeorgiaTechBelinda Li @belindazli
2K Followers 575 Following PhD student @MIT_CSAIL | formerly SWE @facebookai, BS'19 @uwcse | NLP, MLVitor Falcão @vitorfalcaor
18 Followers 376 FollowingSamsudeen @wattosamsu
4 Followers 56 FollowingYarden As @yarden_as
68 Followers 362 FollowingPeter Hase @peterbhase
2K Followers 691 Following Google PhD Fellow at @uncnlp. Interested in interpretable ML, natural language processing, AI Safety, and Effective Altruism.Alberto Fuentes (e/ac.. @AlberFuen
371 Followers 2K Following Cofounder of @daertml. Training LLaMAs as a hobby (and no profit yet).Tyne宇 @Tyne03720826082
110 Followers 3K Followingjinzhuan @jinzhuan2
26 Followers 99 FollowingSergio Soage @Sergio_Soage
904 Followers 5K Following artificial intelligence, math. Random stuff @ https://t.co/tqV9OIPsWEYoung @younqchan
170 Followers 3K Following Final year Ph.D. student working on Out-of-Distribution Generalization and Causality of Large Pre-trained Models, and Graph Neural Networks.Ryan David Cotterell @ryandcotterell
9K Followers 1K FollowingTal Linzen @tallinzen
16K Followers 893 Following Professor @nyuling and @NYUDataScience, research scientist @GoogleAIJosef Valvoda @ValvodaJosef
674 Followers 1K Following PhD candidate @CambridgeNLP group @Cambridge_UniMrinmaya Sachan @mrinmayasachan
2K Followers 2K Following Assistant Professor of Computer Science at ETH Zurich working in natural language processing (#NLProc), machine learning and education (#edtech).Kumar Shridhar @JupyterAI
584 Followers 1K Following PhD in ML/NLP @eth_en | I do #NLProc and #AI | Past: @MSFTResearch @AIatMeta @AmazonScience @rptu_kl_ld | He/him. Views are my own.Tiago Pimentel @tpimentelms
1K Followers 248 Following Postdoc at @ETH_en. Formerly, PhD student at @Cambridge_Uni.Manoel @manoelribeiro
3K Followers 1K Following CS PhD student @ EPFL — On the job market for 2023-2024! 🐘: @[email protected] Keywords: Computational Social Science, Platforms, Communities, ModerationSebastian Gehrmann @sebgehr
5K Followers 2K Following Head of NLP, CTO office, @Bloomberg. (he/him) Generating natural language, one word at a time. Also making sense of that language afterwards. views my ownSam 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.Clara Isabel Meister @clara__meister
2K Followers 49 Following PhD student in the ML Institute at ETH Zurich. Still figuring out how Twitter works... 🤦♀️Christine de Kock @christinedekock
258 Followers 254 Following Researcher & lecturer in NLP at @unimelbSebastian Ruder @seb_ruder
80K Followers 1K Following Multilingual LLMs @cohere • Prev: @GoogleDeepMind • Newsletter: https://t.co/7JGh2qpG98Machel Reid @machelreid
2K Followers 1K Following Research Scientist @GoogleDeepMind Working on LLMs on the Gemini Team; did gemini 1.5 proChristopher Manning @chrmanning
127K Followers 116 Following Director, @StanfordAILab. Assoc. Director, @StanfordHAI. Founder, @stanfordnlp. Prof. CS & Linguistics, @Stanford. IP @aixventureshq. 🇦🇺 Do #NLProc & #AI. 👋ETH Zurich @ETH_en
93K Followers 552 Following This is the official Twitter account of ETH Zurich in English. Stay tuned for the latest news on research, technology and education. Deutscher Account: @ethSebastian Riedel (@ri.. @riedelcastro
15K Followers 470 Following Researcher in NLP/ML @deepmind, @ucl_nlp, @[email protected] on MastodonYoav 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/f9QkzO5kaCEMNLP 2024 @emnlpmeeting
12K Followers 41 Following EMNLP 2024 - The 2024 Conference on Empirical Methods in Natural Language Processing, November 12 –16, 2024 Hashtag: #EMNLP2024Javier Ferrando @javifer_96
277 Followers 480 Following PhD Student @la_UPC. Interpretability in NLPVésteinn Snæbjarnar.. @vesteinns
134 Followers 421 Following PhD fellow at @AiCentreDK @ELLISforEurope @DIKU_institut @BelongieLab • NLP & CVTim Davidson @im_td
537 Followers 396 Following PhD research @EPFL on reliable magic | machine learning & company building | @nyuniversity @UvA_Amsterdam alumn | mostly tweet about AI + some wild thoughtsClément Dumas @Butanium_
78 Followers 217 Following CS MSc student at ENS Paris-Saclay Ecosystem simulation enjoyer/Aspiring AI safety researcherHannah Rose Kirk @hannahrosekirk
3K Followers 684 Following AI researcher trying to make sense of all things cyberspace 🤖 Uni of Ox PhD (loading…) @oiioxford & @OxfordAI. Prev @turinginst & @Cambridge_Uni. Visitor @ NYUEdoardo Debenedetti @edoardo_debe
749 Followers 2K Following CS PhD student @CSatETH 🇨🇭 | ML Security and Privacy 💻🕵️♂️ | prev @EPFL_en @PoliTOnews | Help 🇺🇦 on https://t.co/32YZoUP39z | From 🇪🇺🇮🇹Neil Houlsby @neilhoulsby
4K Followers 318 Following Professional AI researcher; amateur athlete. Senior Staff RS in the Google Deepmind, Zürich. Attempts triathlons.Veniamin Veselovsky @VminVsky
296 Followers 528 Following pre-trained on epfl and university of torontoAnjalie Field @anjalie_f
3K Followers 435 Following @[email protected] Faculty at Johns Hopkins @JHUCompSci @jhuclsp in NLP and computational social scienceImanol Schlag @ImanolSchlag
9 Followers 6 FollowingOlivier Bachem @OlivierBachem
3K Followers 305 Following Senior Staff Research Scientist at @GoogleDeepMind where I lead the team that built the RLHF technology used in Bard, PaLM 2, Gemini, and other Google products.Florian Egli @floegli
1K Followers 584 Following Prof @TU_Muenchen | Researcher @ETH_EPG | Fellow @IIPP_UCL | Associate @foraus | Board @youngacademy_ch | @Monocle24 @GlobalShapers @eth_energy_blogPranav Goel @Pranav__Goel
634 Followers 2K Following Computational social science postdoc at Lazer Lab, Northeastern UniversityPatrick Y. Wu @PatrickYWu
467 Followers 1K Following postdoc @CSMaP_NYU | computational social scientist working on AI+NLP | PhD/MA @UMich, BA @UChicago | https://t.co/yjLeBlS86oAmr Khalifa @AmrMAlameen
823 Followers 393 Following I do AI research and other cool stuff @Google, Google DeepMind team @DeepMind, also PhD student @Mila_Quebec | Opinions are my own |Peter Hase @peterbhase
2K Followers 691 Following Google PhD Fellow at @uncnlp. Interested in interpretable ML, natural language processing, AI Safety, and Effective Altruism.Yarden As @yarden_as
68 Followers 362 Followingjordiae @jordiae
982 Followers 2K Following Transformers, NLP, ML4Code, HPC. PhD student @EdinburghUni. Previously @Bloomberg @MILAMontreal @BSC_CNS @la_upc. Opinions are my own.Zhāng, Miǎo 张淼 @Miao_Zhang_dr
762 Followers 723 Following Post-doc at @cl_uzh doing corpus phonetics. SW Mandarin, Changsha Xiang, Mandarin, Japanese, English, Korean, German, Ikema. He/him. https://t.co/3GwrjPl0v3Christopher Barrie @cbarrie
3K Followers 2K Following Lecturer in Computational Sociology @EdinburghUni @uoessps. Incoming Asst. Prof. @nyuniversity (2024)Corinne Bara @BaraCorinne
1K Followers 644 Following Senior Researcher at ETH Zürich. Interested in the dynamics of violence and strategies of armed actors during civil wars. 🇨🇭🇸🇪Ece Takmaz @ecekt2
883 Followers 2K Following Postdoc at @UniUtrecht, previously PhD candidate at @UvA_AmsterdamXiaohua Zhai @XiaohuaZhai
3K Followers 208 Following Senior Staff Researcher @GoogleDeepMind team in ZürichLivia I. Schubiger @liviaisabella13
3K Followers 2K Following Assoc Prof @Politics_Oxford & @NuffieldCollege; research on conflict, repression, violence.Paul Röttger @paul_rottger
2K Followers 455 Following Postdoc @MilaNLProc, working on evaluating and improving LLM safety. Previously PhD @oiioxford & CTO/co-founder @rewire_onlineAleph Alpha @Aleph__Alpha
7K Followers 2 Following Our mission is a European generalizable AI. We're hiring: https://t.co/TSKL1fbwe0 #AGI, #artificialintelligence, #writtenbyahuman,#writtenbyanAIKiho Park @KihoPark_
292 Followers 105 Following @UChicago Stat PhD student advised by @victorveitchLukas Breitwieser @breitware
11 Followers 12 FollowingMason Meyer @masonmeyer_
232 Followers 126 Following research @openai. but ever with the eternal goal of the true, the beautiful, and the good.Max Schwarzer @max_a_schwarzer
942 Followers 282 Following Doing research at @OpenAI. Did my PhD with Aaron Courville and @marcgbellemare at @Mila_Quebec. Interned at @Apple, @DeepMind, Google Brain, @Numenta.Katherine Lee @katherine1ee
6K Followers 931 Following understanding ourselves and our models. senior research scientist @GoogleBrain, @genlawcenter and @CornellCIS, formerly @Princeton @[email protected]Reza Ghazinouri Ⓥ @GhazinouriEN
377 Followers 395 Following Security Program Manager @GooglePlay. Former co-director @United4Iran. Advocate for rights of human & non-human animals ♀️✊🏿Ⓥ فارسی @ghazinouri Opinions my ownEric Wallace @Eric_Wallace_
6K Followers 1K Following Researcher at OpenAI working to make language models more trustworthy, secure, and private.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.Nick Pangakis @nick_pangakis
252 Followers 329 Following Data Science | Machine Leaning | NLP | PhD Candidate @PennCLS @ChengleiSi
2K Followers 3K Following vibing @stanfordnlp | real AGI is the friends we made along the waymars huang @MarsScHuang
307 Followers 224 Following PhD student at Stanford’s Center for Artificial Intelligence in Medicine & Imaging (AIMI)Ken Liu @kenziyuliu
448 Followers 778 Following CS PhD @StanfordAILab. Thinks about ML privacy, security, localization, trustworthiness. Prev @SCSatCMU, @GoogleAI, @Sydney_Uni 🇦🇺Weiyan Shi @shi_weiyan
3K Followers 694 Following Postdoc @StanfordNLP, incoming assistant professor @Northeastern, PhD @Columbia| Prev Intern @MetaAI |Co-created CICERO | persuasive chatbots + privacy #nlprocOmar Shaikh @oshaikh13
577 Followers 798 Following CS Ph.D. student @Stanford - previously @GeorgiaTech - also @[email protected]Matthias Gerstgrasser @MGerstgrasser
31 Followers 25 Following I teach AIs to play nice with one another, and sometimes with us humans.[1/7] 🚀 Introducing the Language Model Transparency Tool - an open-source interactive toolkit for analyzing Transformer-based language models. We can't wait to see how the community will use this tool! github.com/facebookresear…
The best work I've done has felt like play. I get almost a giddy excitement from new ideas. There is nothing better than working with good people who share your excitement. If you can find a place where work feels like play, you're very lucky. 4/10
There is no globally-optimal life. There is no sequence of choices in life that will produce the "perfect life" or "perfect career". This is hard to accept but, once you accept it, it's very freeing. 2/10
Cool! Reminds me of x.com/debjitpaul2/st… where we use causal analysis to show that LLMs don’t rely on CoT. @sleepinyourhat et al.’s findings align: LLMs use the extra compute during CoT (or here “…”) to babble like a weasely politician while thinking about the real answer…
Will your paper catch the eye of @_akhaliq? I built a demo that predicts if AK will select a paper. It has 50% F1 using DeBERTa finetuned on data from past year. As a test, our upcoming WildChat arXiv has a 56% chance. Hopefully not a false positive🤞 🔗huggingface.co/spaces/yuntian…
The latest talk at @zurichnlp also exists as a video presentation since this morning. 🌂 I welcome feedback on the format, what works, and what doesn't. 🙏 youtu.be/yeEZpf4BlDA
Today I was visiting @WorldPopProject at @unisouthampton. Thank you for the great discussions, I learned a lot today! See you soon!
Launching @SiuuuAI
Advance notice for demo — how a writer might use Siuuu.AI's "Story Writing" feature! #SiuuuAI #AIInnovation #AIWriting
@ninoscherrer Thanks Nino!! Your comments were very helpful 🤩
@hannahrosekirk Still impressed by all the plotting work you did 😅
A very impressive human preference dataset (including annotator demographics) that will without doubt lead to many interesting studies! Congrats on the massive effort @hannahrosekirk!!
Today we're launching PRISM, a new resource to diversify the voices contributing to alignment. We asked 1500 people around the world for their stated preferences over LLM behaviours, then we observed their contextual preferences in 8000 convos with 21 LLMs arxiv.org/abs/2404.16019
This was a nice opportunity to teach and lift some of the magic behind LLM pre-training. I think we got some really cool submissions to our llm-baselines repo: github.com/epfml/llm-base… Thanks a lot to the organizers of @LauzHack for the invitation and organization :)
Transformers Can Represent n-gram Language Models Plenty of existing work has analyzed the abilities of the transformer architecture by describing its representational capacity with formal models of computation. However, the focus so far has been on analyzing the
Published in Nature Machine Intelligence today, our new article explores the trade-offs of personalised alignment in large language models ⚖️ Personalisation has potential to democratise decisions over how LLMs behave, but brings its own set of risks... nature.com/articles/s4225…
Personalised LLMs are great, but should there be limits to personalisation? If so, who should set these limits? For answers to these questions and more, check out our paper on the risks and benefits of personalising LLMs, led by @hannahrosekirk 👇 out in @NatMachIntell today!
Published in Nature Machine Intelligence today, our new article explores the trade-offs of personalised alignment in large language models ⚖️ Personalisation has potential to democratise decisions over how LLMs behave, but brings its own set of risks... nature.com/articles/s4225…
Science has gotten in the way of this account's true purpose. Complaining about einsum.
I've been enjoying Penzai the new Jax lib. It's very opinionated, but close to my ideal NN library. To test it out I ported the Tensor Puzzles to use NamedArrays. Feels so clean without the [:, None]'s srush.github.io/Tensor-Puzzles…
We released 🍷FineWeb: 15T high quality tokens from the web. It's the best ready-to-use AND the largest pretraining dataset. Outperforms all other datasets in our 350B token ablations but scales to much longer training runs due to its sheer size! hf.co/datasets/Huggi…
We have just released 🍷 FineWeb: 15 trillion tokens of high quality web data. We filtered and deduplicated all CommonCrawl between 2013 and 2024. Models trained on FineWeb outperform RefinedWeb, C4, DolmaV1.6, The Pile and SlimPajama!
We've recently released Coix, a JAX-based framework designed for composing probabilistic programs and performing inference on them. Let's make Sequential Monte Carlo easy to apply. Tutorials are here coix.readthedocs.io/en/latest/ Joint work with @zmheiko @tuananhle7 @sharadvikram