Kyunghyun Cho @kchonyc
a combination of a mediocre scientist, a mediocre manager, a mediocre advisor & a mediocre PC at @nyuniversity (@CILVRatNYU) & @genentech (@PrescientDesign). kyunghyuncho.me Manhattan, NY Joined June 2009-
Tweets12K
-
Followers60K
-
Following2K
-
Likes32K
🚨 Iterative Reasoning Preference Optimization 🚨 - Iterative algorithm for reasoning tasks: generate pairs & apply DPO+NLL - Improves accuracy over iterations on GSM8K, MATH, ARC & beats baselines E.g. Llama2-70B GSM8K: 55.6%->81.6% (88.7% maj32) arxiv.org/abs/2404.19733 🧵(1/5)
i use managed @WordPress from @EasyWP via @Namecheap where i get my domain name.
i use managed @WordPress from @EasyWP via @Namecheap where i get my domain name.
Thanks @LangChainAI ! 🦜🔗🌞 Introducing the “langchain-upstage integration package". Explore cutting-edge features for your RAG, such as the groundedness check model and layout analysis(document loader). Our latest integration, including the Solar chat model and text embedding…
a new blog post, because it is Saturday. <Fixing DPO but I have a dinner reservation …> kyunghyuncho.me/a-proper-prefe…
Congrats to @ArtieShen on being honored by @nyuniversity w/ an outstanding dissertation award in the public health & allied health category for "Toward Explainable #DeepLearning for Medical Image Analysis." 🧵 1/10
all … thanks for teaching me lagrangian, etc. lesson learned: use X, Y and Z …
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.
all tiktok needs to do is to add a button that triggers some guns in a desert.
%env ABC="XYZ" vs. export ABC="XYZ" can you guess the difference? 🤦
I got tenure! It was fitting that I got to celebrate with the lab right after the news. Working together for the last 6.5 years has been a blast.
your leadership doesn't want you to do it: have you tried Pubmed-QA zero-shot eval without questions but only with abstracts?
it turned out @AIatMeta 's llama-3 8b is conscious but also useless in some cases 🤣
arxiv.org/abs/2404.08819 a nice study by Merrill, Petty & Sabharwal. it looks like i won't have to wait too much longer for the reinvention of LSTM/GRU by LLM bros.
It's been a wild ride. Just 20 of us, burning through thousands of H100s over the past months, we're glad to finally share this with the world! 💪 One of the goals we’ve had when starting Reka was to build cool innovative models at the frontier. Reaching GPT-4/Opus level was a…
It's been a wild ride. Just 20 of us, burning through thousands of H100s over the past months, we're glad to finally share this with the world! 💪 One of the goals we’ve had when starting Reka was to build cool innovative models at the frontier. Reaching GPT-4/Opus level was a…
i was suddenly reminded of this article from 2018. ".. statistical physicists were primed to see power laws everywhere .. there’s a “power law religion.”" "“We would even squint at the computer screen from an angle to get a better idea if a curve was straight or not,” recalled…
oh i think i figured out: 1. create a new repository on github 2. clone the repository in @LightningAI studio 3. press "Open" to work directly on the cloned repo
oh i think i figured out: 1. create a new repository on github 2. clone the repository in @LightningAI studio 3. press "Open" to work directly on the cloned repo https://t.co/ZUhSxDy0Fx
(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K FollowingPercy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistSasha Rush @srush_nlp
52K Followers 464 Following Professor, Programmer in NYC. Cornell Tech, Hugging Face 🤗 https://t.co/cZl0wTfqGzGautam Kamath @thegautamkamath
44K Followers 507 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Rosanne 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/IbycyGfnDRJia-Bin Huang @jbhuang0604
51K Followers 285 Following Associate Professor @umdcs; Part-time Research Scientist @Meta. I like pixels.NeurIPS Conference @NeurIPSConf
112K Followers 35 Following New Orleans, Dec 10-16, 23. https://t.co/ga8aOw615g Tweets to this account are not monitored. Please send feedback to [email protected].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)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.Yi Tay @YiTayML
29K Followers 97 Following chief scientist / cofounder @RekaAILabs 🫠 past: research scientist @google brain 🤯 currently learning to be a dad 🍼Graham Neubig @gneubig
31K Followers 588 Following Associate professor at CMU, studying natural language processing and machine learning.Zachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷yobibyte @y0b1byte
15K Followers 2K Following Kurin ViTaly, senior research scientist @IsomorphicLabs, ML PhD from @UniofOxford on RL, Multitask learning & GraphsBehnam Neyshabur @bneyshabur
18K Followers 690 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingAkari Asai @AkariAsai
11K Followers 650 Following Ph.D. student @uwcse & @uwnlp. NLP. IBM Ph.D. fellow (2022-2023). Meta student researcher (2023-) . ☕️ 🐕 🏃♀️🧗♀️🍳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'sEdward Grefenstette @egrefen
36K Followers 776 Following FR/US/GB AI/ML Person, Director of Research at @GoogleDeepMind, Honorary Professor at @UCL_DARK, @ELLISforEurope Fellow. All posts are personal.Sander Dieleman @sedielem
50K Followers 2K Following Research Scientist at Google DeepMind. I tweet about deep learning (research + software), music, generative models (personal account).mmm @mmm185939
0 Followers 2 FollowingBray🤘🏽 @bravious_wooten
0 Followers 120 FollowingBoris @ToushIFHAQ
34 Followers 259 Following Made in India Believe in yourself Above all have faith in AllahMargaret Oluwatoyin s.. @margaret_sippio
0 Followers 99 Following Being an entrepreneur,is a special part of my life,that allows me connect with humanity,see every moment as an opportunity to learn,explore,discover, and teach,jenny_yiong @Jennyyiong14
2 Followers 164 Following Beauty, cosmetics & personal care Fairies rely on faces to make a living Girls refuse to admit defeat👧ᵕ̈ ᑋᵉᑊ 🔸ℚ𝔹𝔼𝔼𝔽𝕃𝕐 Brand|Quick consultationGaB @THFC_GaB
412 Followers 864 Following Spurs since 1972! ST holder Paxton Block 515. ST Ulster Rugby #COYS #Stones Kentish man living in Belfast. Support #autismXinyu Zhao @lucy_xyzhao
1 Followers 32 Following. @mysticmelt
11 Followers 20 FollowingCrypto Hawk 🦅 @CryptoHawk_07
80 Followers 578 Following #ALTCOIN Gem Hunter #Crypto marketing expert #Crypto Genius 💰My Post’s are NFA # BTC #BNB #ETH 💵 DM for Promotion 📥🙌Hardy Ilunga @ilunga_hardy
13 Followers 177 Following Étudiant à l'Institut supérieur des techniques appliquées de Lubumbashian zhang @anzhang1122172
0 Followers 22 FollowingBALDEZO @Baldezo1004
137 Followers 979 Following 🗣️Soit à l'écoute, tout parle, tout est parole, tout cherche à nous communiquer une connaissance ✨⚡Sharon Owino @5b662c44eef949c
2 Followers 23 FollowingYamei Chen @NOrangeroli
6 Followers 129 FollowingXinbao Qiao @Xinbao_Qiao
0 Followers 8 FollowingGeorge Smith @georgeksmith
298 Followers 465 Following I really don't want to be addicted to Twitter, so I post randomly and rarely.ndjdbdud @WuWuming
98 Followers 449 FollowingMichiel van de Panne @Mvandepanne
4K Followers 416 Following UBC Computer Science; physics-based models of human movement; deep reinforcement learning; animation; robotics. Connect with me on that other network.Paylz @paylza
139 Followers 2K Following The best online market for digital downloads with best prices.Educarte IA @EducarteIa
268 Followers 3K Following Desarrollador de soluciones con inteligencia artificial / Consultor Bussines IA / Researcher IA / especialista en SEO / Formulador de proyectosDaniel Zheng @dzheng256
46 Followers 352 FollowingF.Mackenzie @mackenzie85372
7 Followers 142 FollowingHaniwa@営業DXコン.. @consulting_dx
8 Followers 71 Following 技術顧問/Web・アプリ開発/DXコンサル/生成AI/エンジニア #SalesforceAnirudh Atmakuru @aatmakuru6
5 Followers 50 FollowingChad @al0k0
156 Followers 1K Following Insignificant, Self conscious, carbon based life-form. 13.772 billion light years old.sweetnightmare @chin__20
27 Followers 136 FollowingOpen @OpenXuu
0 Followers 150 Followingcarpedm30 @carpedm30
4K Followers 443 Followingsaid hammoudi @mohsaid_pro
20 Followers 101 FollowingZhaoyang Chu @zhaoyang_c68411
9 Followers 365 Following CS Master@HUST. Interested in SE+ML, specifically focusing on building trustworthy and reliable AI-based software systems. Seeking PhD starting in 2025 Fall.Daniel Levi-Minzi @dlvmnz
49 Followers 149 Following@pretoria2012 @pretoria2012
106 Followers 1K FollowingAli Kheirandish @AliKheirandish4
156 Followers 4K FollowingJose Finger @jvfinger
252 Followers 3K Following SQL Server, Oracle, Data warehouse, Alteryx, Tableau. Dogs and Cats. Warhammer 40k Fan.Automation for farmsenming @EnmingYuan
15 Followers 149 FollowingJ Sam🌐 @JaicSam
540 Followers 5K Following #LifestyleMedicine Doctor #AI #ML #DataScience Masters in #Philosophy Memes ≠ medical adviceDavid Vivancos - e/ac.. @VivancosDavid
1K Followers 1K Following Teaching Machines | Advising Human CEOs | Building Neurotechnologies | Opening Events Beyond AI | Read my books "The End of Knowledge & Automate or Be AutomatedNguyễn Cường @1002cuong_me
5 Followers 57 FollowingRamalingam @ramlingamr
140 Followers 4K Following Reader; I share items that interest me in AI, Skills / Talent, Healthcare, Investments and Chess.(((ل()(ل() 'yoav))).. @yoavgo
46K Followers 2K FollowingPercy Liang @percyliang
49K Followers 408 Following Associate Professor in computer science @Stanford @StanfordHAI @StanfordCRFM @StanfordAILab @stanfordnlp | cofounder @togethercompute | PianistLucas 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]Sasha Rush @srush_nlp
52K Followers 464 Following Professor, Programmer in NYC. Cornell Tech, Hugging Face 🤗 https://t.co/cZl0wTfqGzGautam Kamath @thegautamkamath
44K Followers 507 Following Assistant Prof of CS @UWaterloo, Faculty @VectorInst, Canada @CIFAR_News AI Chair. Co-EiC @TmlrOrg. I lead @TheSalonML. Privacy, robustness, machine learning.Rosanne 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/IbycyGfnDRJia-Bin Huang @jbhuang0604
51K Followers 285 Following Associate Professor @umdcs; Part-time Research Scientist @Meta. I like pixels.NeurIPS Conference @NeurIPSConf
112K Followers 35 Following New Orleans, Dec 10-16, 23. https://t.co/ga8aOw615g Tweets to this account are not monitored. Please send feedback to [email protected].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)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.Christopher Manning @chrmanning
127K Followers 116 Following Director, @StanfordAILab. Assoc. Director, @StanfordHAI. Founder, @stanfordnlp. Prof. CS & Linguistics, @Stanford. IP @aixventureshq. 🇦🇺 Do #NLProc & #AI. 👋Yi Tay @YiTayML
29K Followers 97 Following chief scientist / cofounder @RekaAILabs 🫠 past: research scientist @google brain 🤯 currently learning to be a dad 🍼Graham Neubig @gneubig
31K Followers 588 Following Associate professor at CMU, studying natural language processing and machine learning.Zachary Lipton @zacharylipton
59K Followers 2K Following Professor: CMU/@acmi_lab, CTO / CSO: @AbridgeHQ, Creator: @d2l_ai & https://t.co/QQt98VNLUp, Relapsing 🎷yobibyte @y0b1byte
15K Followers 2K Following Kurin ViTaly, senior research scientist @IsomorphicLabs, ML PhD from @UniofOxford on RL, Multitask learning & GraphsBehnam Neyshabur @bneyshabur
18K Followers 690 Following Senior Staff Research Scientist @GoogleDeepMind, Interested in reasoning w. LLMs, traveling & backpackingAkari Asai @AkariAsai
11K Followers 650 Following Ph.D. student @uwcse & @uwnlp. NLP. IBM Ph.D. fellow (2022-2023). Meta student researcher (2023-) . ☕️ 🐕 🏃♀️🧗♀️🍳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'sBob West @cervisiarius
2K Followers 113 Following Associate Professor at EPFL, Data Science Lab (dlab)Rafael Rafailov @rm_rafailov
3K Followers 637 Following Ph.D. Student at @StanfordAILab. I work on Foundation Models and Decision Making. Previously @GoogleDeepMind @UCBerkeleyHugh Zhang @hughbzhang
1K Followers 523 Following open source ai @scale_AI. co-created @gradientpub.Haresh Rengaraj @HareshRengaraj
35 Followers 581 FollowingStefano Martiniani @SteMartiniani
1K Followers 2K Following Assistant Professor of Physics, Chemistry, and Mathematics at NYU | @SimonsFdn Faculty Fellow | @nyu_csmr @SimonsCenterNYU | @Gates_Cambridge AlumOded Regev @regevlab
430 Followers 13 FollowingSam Power @sp_monte_carlo
17K Followers 7K Following Lecturer in Maths & Stats at Bristol. Interested in probabilistic + numerical computation, statistical modelling + inference. (he / him)Howard Chen @__howardchen
854 Followers 1K Following PhDing @princeton_nlp & @PrincetonPLI. Previously: Meta AI (intern) / ASAPP research / Cornell Tech / NTU (Taiwan).Git Maxd @GitMaxd
652 Followers 425 Following Linux Dev | AI Rev Early Adopter | Love React | Content Creator | Founder 3x startups (+$15mm)Bill Byrne @unattributed
99 Followers 563 Following speech and language processing researcher, also machine learning, professor, emigrant, bicyclistDavid K. Yang @davidkmyang
1K Followers 2K Following Life sciences @Lux_Capital. prev. @8VC @Sabeti_Lab @Facebook @BroadInstituteAlbert Jiang @AlbertQJiang
2K Followers 409 Following AI4Maths @Cambridge_CL Science @MistralAI I bake my own opinions at temperature=2.0Hwaran Lee @hwaran_lee
255 Followers 186 Following Lead Research Scientist @NAVER_AI_Lab | PhD from @KAISTAaron Defazio @aaron_defazio
6K Followers 365 Following Research Scientist at Meta working on optimization. Fundamental AI Research (FAIR) teamInner City Press @innercitypress
267K Followers 3K Following Matthew Russell Lee for/as Inner City Press covers SDNY, UN Gate, banks & IMF. books https://t.co/xHL0pGID4n https://t.co/VTEqaLISDBElad Hazan @HazanPrinceton
11K Followers 187 Following machine learning and optimization @PrincetonCS & Google DeepMind Princeton, dad^3Peter Stone @PeterStone_TX
2K Followers 238 Following Prof. of Computer Science at UT Austin with research interests in AI, robotics, machine learning, multiagent systems; Executive Director of Sony AI, AmericaWuming Gong @WumingG
73 Followers 357 Following Principal Scientist at Genentech | Assistant Professor at University of MinnesotaPaul Ohm @paulohm
6K Followers 430 Following @GeorgetownLaw Professor / Computer Programmer / Former Computer Crime Prosecutor and Former FTC Senior Policy AdvisorDavid Hall @dlwh
2K Followers 1K Following Research Engineering Lead at @StanfordCRFM . Previously co-founder at Semantic Machines ⟶ MSFT. Lead developer of Levanter, Breeze. he/him @[email protected]noahdgoodman @noahdgoodman
2K Followers 109 Following Professor of natural and artificial intelligence @Stanford. Research Scientist at @GoogleDeepMind. (@StanfordNLP @StanfordAILab etc)Francesco Orabona @bremen79
6K Followers 394 Following Associate professor at @KAUST_News. Formerly @BU_ece, @sbucompsc, @YahooResearch, @TTIC_Connect. ML theory&practice and history of scienceJinwoo Leem @ideasbyjin
962 Followers 584 Following Director of ML @alchemabtx prev: @benevolent_ai, @OPIGlets. Antibodies, LLMs, protein structure prediction, neuro, onco. 🇨🇦🇬🇧🇰🇷 Tweets=my views.Nicholas Lourie @NickLourie
151 Followers 313 Following I build things. 🤖 Doing a PhD at @nyuniversity (@CILVRatNYU) on better empirical methods for deep learning and data science. Advised by @kchonyc and @hhexiy.Michal Valko @misovalko
5K Followers 2K Following Llama @AIatMeta Paris & Inria & MVA - Ex: Gemini and BYOL @GoogleDeepMindSungmin Cha @_sungmin_cha
53 Followers 63 Following Faculty Fellow @nyuniversity | PhD @SeoulNatlUniいさご@LINE_AI_Com.. @shin135
8K Followers 2K Following LINE AIカンパニーCEO、AI事業統括担当執行役員。LINEがもつAI技術の社会実装を推進しています。Clova組み込み、チャットボット、音声認識、音声合成、OCRなど日本語や日本のコンテンツに特化したAI技術取り揃えておりますのでお気軽にご相談くださいませ。引き続きDeveloper Relationsも担当。Ahmad Beirami @abeirami
4K Followers 2K Following Building safe, helpful, and scalable generative AI @Google | ex-{@AIatMeta, @EA, @MIT, @Harvard, @DukeU} | @GeorgiaTech PhD | زن زندگی آزادی | opinions my ownGroq Inc @GroqInc
46K Followers 470 Following Creator of the LPU™ Inference Engine, providing the fastest speed for AI applications, designed & engineered in N. America https://t.co/DsEqVAC5DpIdeogram @ideogram_ai
39K Followers 0 Following Helping people become more creative. It's pronounced eye-diogram. Join our lovely community at https://t.co/aKDNl4OOQf.Katie Bouman @klbouman
4K Followers 0 Following Assistant Professor of CMS/EE @Caltech; Computational Imaging, Computer Vision, and Machine LearningBioptimus @bioptimus_ai
400 Followers 2 Following We build foundation models that will transform biology.AndriyMulyar @andriy_mulyar
11K Followers 517 Following building tech that enables humans to interact with latent spaces 🗺️ founder / cto @ https://t.co/NbsLHLWfy8 prev. ML Ph.D. Student at NYU CourantJiacheng Liu (Gary) @liujc1998
991 Followers 188 Following 🎓 PhD student @uwcse @uwnlp. 🛩 Private pilot. Previously: 🧑💻 @oculus, 🎓 @IllinoisCS. 📖 🥾 🚴♂️ 🎵 ♠️Satnam Singh @satnam6502
14K Followers 3K Following Punjabi-Scottish-American Haskell hacker at @GroqInc, cook, cyclist, lost in music. ∃🇮🇳 ∧ ∀🇬🇧 ∧ ∃🇪🇺 ∧ ∀🇺🇸 #celiac ex-{Microsoft, Google, Facebook}Niloofar (Fatemeh) Mi.. @niloofar_mire
4K Followers 1K Following Postdoc @uwcse-@uwnlp, Ph.D. from @ucsd_cse /Privacy, ML, NLP, @winlpworkshop chair, @MSFTResearch - Semantic MachinesDavid Barber @davidobarber
3K Followers 741 Following Director UCL Centre for AI and UiPath Distinguished Scientist. Co-founder https://t.co/Wx3VpUByR2. Pro: cycling, walking, EU. @[email protected]. Views my own.Teresa Datta @teresa_datta
37 Followers 88 Following transparency & social impact of AI from a human-centered lens | ML @itsArthurAIAnton Tsitsulin @tsitsulin_
2K Followers 420 Following Graphs, ML, math, data. Research Scientist @GoogleAI (not the DeepMind kind). Ph.D. from @UniBonn.Yanli Zhou @yanlisagezhou
101 Followers 69 Following PhD student with @LakeBrenden at @NYUDataScienceNathan Lambert @natolambert
25K Followers 690 Following Figuring out AI @allen_ai, "rl boi" DM me papers. Writes @interconnectsai, talks @retortai Has phd and some credentialsAlaa El-Nouby @alaa_nouby
522 Followers 302 Following Research Scientist at @Apple. Previous: @Meta (FAIR), @Inria, @MSFTResearch, @VectorInst and @UofG . Egyptian 🇪🇬 Deprecated twitter account: @alaaelnoubyKeith Hornberger @KRHornberger
11K Followers 288 Following Exec. director, medicinal chemistry @ArvinasInc | PhD @Columbia | hiker | beer & cocktail enthusiast | he/him/huz/dad | dreaming of @SedonaAZ | all posts my ownKenneth Church @kchurch4
98 Followers 45 FollowingThe Civil Rights Lawy.. @johnbryanesq
29K Followers 893 Following WV Civil rights Lawyer, Youtuber and Unlicensed Historian and Scavenger. Freedom is scary. https://t.co/4csuauOxd1every year around now i’m frustrated by the gross inefficiency (economic sense) of the CS faculty job market. it works-ish for the top 1% of departments and candidates and is awful for others. but without buyin from that 1% of departments it’s enormously difficult to change.
Results on ARC and MATH give similar positive conclusions to GSM8K. We see improvements over iterations on all 3 tasks, which tend to diminish in iteration 3-4. Overall, Iterative RPO gives strong results given the base model & not using extra data. Thanks for reading! 🧵(5/5)
Negative examples are also crucial, as SFT tends to assign similar probability to chosen and rejected generations from our DPO pairs (see fig). DPO+NLL fixes this, and beats SFT in task accuracy (73.1% on iteration 1 vs. 63.5%). 🧵(4/5)
We find the NLL term is crucial, see GSM8K results in 1st tweet (73.1% vs. 61.8%). One reason: without this the log prob of chosen examples in DPO decreases (see fig). Note: chosen answers are known to be correct, hence NLL makes more sense here than in other DPO setups 🧵(3/5)
Recipe 👩🍳: Start with base model & fixed training set with labels. - Generate multiple CoTs + answers per train example with current model - Build preference pairs based on answer correct vs. not - Train DPO + NLL term (for correct answers) Repeat steps with new model 🧵(2/5)
🚨 Iterative Reasoning Preference Optimization 🚨 - Iterative algorithm for reasoning tasks: generate pairs & apply DPO+NLL - Improves accuracy over iterations on GSM8K, MATH, ARC & beats baselines E.g. Llama2-70B GSM8K: 55.6%->81.6% (88.7% maj32) arxiv.org/abs/2404.19733 🧵(1/5)
Tomorrow is my last day at @inflectionAI. What a great ride! Some highlights in this THREAD. 1/
never thought i'd be featured in the financial times, but if i am going to be featured, i'm glad it's because of my crazy commute! 😂 ft.com/content/26b552…
Glad the marathon of faculty interview season is drawing to a close. I had a dream Taylor Swift was visiting UWaterloo and I signed up for a 30 minute slot in her schedule. Unfortunately all our meeting time was eaten up because we couldn't find where she left her backpack.
@haldaume3 I got it immediately below this post! But the other ad I keep getting is the one from Anthropic where they pivot to being “the one enterprises trust”. The xrisk to IBM pipeline is strong 🤷♂️
i now get paid ads to learn about xrisk what a world
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…
Professor life is off to a great start! Honored to receive a grant from Apple ML Research and to be named a Google Research Scholar. Looking forward to more work developing ML methods for healthcare and equity Pictured: an apple, Google, and me
all it says to me the architectures and models are becoming more and more similar, rather than dataset being everything. good luck training neural nets without residual connections.
The dataset is everything. Great read: nonint.com/2023/06/10/the…
@kchonyc @stanfordnlp Thanks for sharing these thoughts, @kchonyc, very interesting! Small typo in this equation (the two p-terms snuck in but shouldn’t have):
🧐🧐🧐🧐 the timing man, the timing
Microsoft has open sourced MS-DOS under MIT license
@wellingmax @kchonyc Yeah, I came up with this many years ago, though I wrote it F(X)=0. Every theory is a special case of this. 🤪Hence every theory paper should cite me 🤪🤣. yann.lecun.com/ex/fun/index.h… [there are researchers who make similar claims in real life without joking. I won't name names 😇 ]
@kchonyc @wellingmax @ylecun elaborate "dog ate my homework"
Dr. Shen's dissertation was advised by @kjgeras, assistant professor of radiology at @nyugrossman, & @kchonyc, professor of computer science at @NYU_Courant. Dr. Shen, who goes by Artie, is now an assistant professor of #radiology at NYU Grossman: med.nyu.edu/faculty/yiqiu-… 6/10