Aditya Arun @adityaarun1
Minimalist, Research Scientist @ Adobe, PhD adityaarun1.github.io Noida, India Joined March 2010-
Tweets8K
-
Followers345
-
Following1K
-
Likes21K
PPO vs GRPO vs REINFORCE – a workflow breakdown of the most talked-about reinforcement learning algorithms ➡️ Proximal Policy Optimization (PPO): The Stable Learner It’s used everywhere from dialogue agents to instruction tuning as it balances between learning fast and staying…
🆕 𝐔𝐩𝐝𝐚𝐭𝐞!! A few additional findings for 𝐑𝐨𝐥𝐥𝐨𝐮𝐭–𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐌𝐢𝐬𝐦𝐚𝐭𝐜𝐡: ① 𝐏𝐚𝐫𝐚𝐥𝐥𝐞𝐥𝐢𝐬𝐦 𝐝𝐢𝐟𝐟𝐞𝐫𝐞𝐧𝐜𝐞 is a huge driver of the gap, with Sequence Parallelism (SP) causing most high max mismatch. ② 𝐋𝐨𝐧𝐠𝐞𝐫 𝐬𝐞𝐪𝐮𝐞𝐧𝐜𝐞𝐬…
🆕 𝐔𝐩𝐝𝐚𝐭𝐞!! A few additional findings for 𝐑𝐨𝐥𝐥𝐨𝐮𝐭–𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐌𝐢𝐬𝐦𝐚𝐭𝐜𝐡: ① 𝐏𝐚𝐫𝐚𝐥𝐥𝐞𝐥𝐢𝐬𝐦 𝐝𝐢𝐟𝐟𝐞𝐫𝐞𝐧𝐜𝐞 is a huge driver of the gap, with Sequence Parallelism (SP) causing most high max mismatch. ② 𝐋𝐨𝐧𝐠𝐞𝐫 𝐬𝐞𝐪𝐮𝐞𝐧𝐜𝐞𝐬… https://t.co/3VxvsPfm8M
🚀 We're thrilled to unveil Qwen3-VL — the most powerful vision-language model in the Qwen series yet! 🔥 The flagship model Qwen3-VL-235B-A22B is now open-sourced and available in both Instruct and Thinking versions: ✅ Instruct outperforms Gemini 2.5 Pro on key vision…
Vision tokenizers are stuck in 2020🤔while language models revolutionized AI🚀 Language: One tokenizer for everything Vision: Fragmented across modalities & tasks Introducing AToken: The first unified visual tokenizer for images, videos & 3D that does BOTH reconstruction AND…
Vision tokenizers are stuck in 2020🤔while language models revolutionized AI🚀 Language: One tokenizer for everything Vision: Fragmented across modalities & tasks Introducing AToken: The first unified visual tokenizer for images, videos & 3D that does BOTH reconstruction AND…
Supplementary information for the new DeepSeek R1 Nature paper is very interesting! Details on training data, hyperparameters, base model importance, and more.
John von Neumann is the father of modern game theory. His 1928 paper and his 1944 book, "Theory of Games and Economic Behavior," co-authored with Oskar Morgenstern, founded the field. He introduced the minimax theorem for zero-sum games, a revolutionary concept showing that…
2025 is 70% complete.
Super excited to share our work on variable-length discrete diffusion models! We design a diffusion process that iteratively unmasks tokens and simultaneously adds new tokens. This enables variable-length discrete planning as well as flexible text generation and reasoning!
Super excited to share our work on variable-length discrete diffusion models! We design a diffusion process that iteratively unmasks tokens and simultaneously adds new tokens. This enables variable-length discrete planning as well as flexible text generation and reasoning!
New diffusion RL method from Tencent:SRPO! Directly Aligning the Full Diffusion Trajectory with Fine-Grained Human Preference 32 H20 with 10minutes enhance Flux 1-dev.
IMO — Fei-fei is spot on... and this is why #SyntheticData for LLMs is a weak strategy for AGI Consider the following example: - Step 1: take a large LLM training dataset (e.g. "the whole internet") - Step 2: remove every sentence with the words "Kim" or "Kardashian" in it…
IMO — Fei-fei is spot on... and this is why #SyntheticData for LLMs is a weak strategy for AGI Consider the following example: - Step 1: take a large LLM training dataset (e.g. "the whole internet") - Step 2: remove every sentence with the words "Kim" or "Kardashian" in it…
Join me this Tuesday for a virtual workshop on Making Poetry Comics from Nature!
Extremely proud of our recent "Lost in Latent Space" paper by our amazing @PolymathicAI intern @FrancoisRozet ! How far do you think we can/should compress ? What is needed in the latent space to keep rolling out the physics right across so many scales?
Extremely proud of our recent "Lost in Latent Space" paper by our amazing @PolymathicAI intern @FrancoisRozet ! How far do you think we can/should compress ? What is needed in the latent space to keep rolling out the physics right across so many scales? https://t.co/IoYFQglL6I
Language models often produce repetitive responses, and this issue is further amplified by post-training. In this work, we introduce DARLING, a method that explicitly optimizes for both response diversity and quality within online reinforcement learning!
Language models often produce repetitive responses, and this issue is further amplified by post-training. In this work, we introduce DARLING, a method that explicitly optimizes for both response diversity and quality within online reinforcement learning!
🍌 nano banana is here → gemini-2.5-flash-image-preview - SOTA image generation and editing - incredible character consistency - lightning fast available in preview in AI Studio and the Gemini API
Claim: gpt-5-pro can prove new interesting mathematics. Proof: I took a convex optimization paper with a clean open problem in it and asked gpt-5-pro to work on it. It proved a better bound than what is in the paper, and I checked the proof it's correct. Details below.
Something that I can totally attest to.
Something that I can totally attest to.
Chain-of-Agents Interesting idea to train a single model with the capabilities of a multi-agent system. 84.6% reduction in inference cost! Distillation and Agentic RL are no joke! Here are my notes:
Evaluating language models is tricky, how do we know if our results are real, or due to random chance? We find an answer with two simple metrics: signal, a benchmark’s ability to separate models, and noise, a benchmark’s random variability between training steps 🧵
Evaluating language models is tricky, how do we know if our results are real, or due to random chance? We find an answer with two simple metrics: signal, a benchmark’s ability to separate models, and noise, a benchmark’s random variability between training steps 🧵 https://t.co/8sAL5yWbh3
New survey on diffusion language models: arxiv.org/abs/2508.10875 (via @NicolasPerezNi1). Covers pre/post-training, inference and multimodality, with very nice illustrations. I can't help but feel a bit wistful about the apparent extinction of the continuous approach after 2023🥲
Today we're putting out an update to the JAX TPU book, this time on GPUs. How do GPUs work, especially compared to TPUs? How are they networked? And how does this affect LLM training? 1/n
Makarand Tapaswi @MakarandTapaswi
1K Followers 632 Following Principal ML Scientist @WadhwaniAI | Assistant Professor @IIIT_Hyderabad | Opinions my own
Dan Roy @roydanroy
57K Followers 2K Following ML / AI researcher. Research Director and Canada CIFAR AI Chair, @VectorInst. Professor, @UofT (Statistics/CS).
Tarun Kalluri @tarun_05
691 Followers 2K Following Research @Meta. PhD @UCSanDiego. Proud Indian 🇮🇳.
Caglar @caglar_ee
9K Followers 5K Following PhD in EEE, Research: machine learning, probabilistic generative models, convex optimization
Tejaswi Kasarla @tkasarla_
2K Followers 617 Following Research Scientist Intern at FAIR Paris @AIatMeta | Learning in non-Euclidean spaces | PhD student @UvA_Amsterdam @Ellis_Amsterdam | Board Member @WiCVWorkshop
Charu Sharma @sh_charu
1K Followers 711 Following Assistant Professor, IIIT Hyderabad, PhD, IITHyderabad | Geometric ML/DL (graphs, point clouds), 3D scene understanding, 3D+LLM
WACV @wacv_official
5K Followers 985 Following The IEEE’s and the PAMI-TC’s premier meeting on applications of computer vision | March 6-10, 2026 Tucson, Arizona
KochenMitFreunden @IBD0e7KEM8aLn
19 Followers 700 Following
Robert Scoble @Scobleizer
543K Followers 24K Following The best from ML/AI community | Ex-Microsoft, Rackspace, Fast Company | Wrote eight books about the future | Silicon Valley robots, holodecks, BCIs, & startups.
Minhyuk Sung @MinhyukSung
2K Followers 635 Following Associate professor @ KAIST | KAIST Visual AI Group: https://t.co/mblvQKFc8t.
Keshav Gupta @keshav0306
3 Followers 126 Following
Mehul Damani @MehulDamani2
598 Followers 405 Following PhD-ing @mit @MIT_CSAIL | language models, reinforcement learning
Kawshik Manikantan @skawshik11
6 Followers 121 Following
Uqinif @Uqinif202
115 Followers 3K Following
jessica🩶 @ds_jessica_
14K Followers 12K Following analytics lead & angel investor & advisor. always learning = business & innovation. doing #datascience
Jereegorv @Jereegorv29224
6 Followers 411 Following
Sathwik @VishnuSathvik1
770 Followers 705 Following Trying to make machines reason @precogatiiith | ug @iiit_hyderabad | Tweets everything.
Omid Nejati @Manzarii
212 Followers 2K Following PhD student @Concordia University | CV and ML researcher | interested in Medical Imaging
Garrison Amaya @HerbieMari24744
1 Followers 98 Following Recruiting webshell engineers to penetrate websites, with a monthly salary of up to $100,000. If interested, please contact https://t.co/aiAe4faFvX
Aadith Warrier @AadithWarrier
21 Followers 341 Following Robotics and Vision | Research @ RRC IIIT-H | MechE @ BITS Pilani '25
Amath SOW @AmathSo36561819
776 Followers 3K Following PhD student at Reasoning and Learning Lab(Real), Linkoping(Sweden): Multi-agent planning in dynamic and uncertain env, RL , meta heuristics, Robotics
Prathyush Sambaturu @prathyushspeaks
2K Followers 515 Following Postdoc @UniofOxford @OxfordBiology @PSIOxford | Ph.D. @uva | MS @iiit_hyderabad | he/him
DaleDunbar @6lZ0V3x7wLYSzo
65 Followers 2K Following
accelerate(e/acc) @EAccelerate_42
3K Followers 7K Following Software,Hardware,Astrophysics, Web3, Stocks and AI enthusiast,In perplexity over answer to life universe and Everything . Founder at https://t.co/R6wDNPGZKK
Stéphane Deny @StphTphsn1
4K Followers 7K Following Neuroscience & ML Researcher. Posting about various topics on here. Tag me for a retweet.
Anri @Anri1014128
27 Followers 281 Following 副業で収入を増やすコツを紹介中。副業を始めたい人にお得な情報をお届け。少しですが10万円プレゼント企画も実施中♪ フォロワーさん優先⇒ https://t.co/WIHx97HG1M
Stotoyez @StotoyezHGHRHh
44 Followers 947 Following
Deepti Rawat @dpt_xyz
20 Followers 114 Following Research Scholar at Centre for Visual Information Technology (CVIT) at IIIT-H, India https://t.co/z5WwKU8f7q
Matthew Macfarlane @MattVMacfarlane
799 Followers 2K Following PhD Candidate @AmlabUvA. Working on Reinforcement Learning, Meta Learning and Reasoning. Prev intern @Microsoft. Views Are Not My Own. 🏴
Priyanka Kargupta @priyanka_karg
213 Followers 318 Following structured nlp to help models & humans critically think phd cand @UIUC_NLP @MSFTResearch | nsf grfp fellow prev: cs'22 @berkeley_ai | @IntelAI | @IBMResearch
Tawffee @TawffeeO2CG3
71 Followers 1K Following
cute @cute91326735897
159 Followers 5K Following
Souvik Ghosh @souvikg_tweets
131 Followers 913 Following Audio Visual AI at CVIT, @iiit_hyderabad | Previously Researcher @synclabs_so and @WhilterAI | Building @cognitiinc |
Heming Zhu @wootwootwo
438 Followers 988 Following Ph.D. student at MPI Informatics @VcaiMpi | Human Digitalization & 3D Reconstruction
Shubham Raj @realrajshubham
95 Followers 55 Following Chief Technology Officer @Kalp_Official | Bringing regulated real world assets on Web3 | Building public-permissioned cross-chain RWA ecosystem
Yasser Benigmim @yasserbenigmim
155 Followers 1K Following PhD student at Télécom Paris, @DeepLearning, @ComputerVision
Michael. @Michael28623704
108 Followers 5K Following
Smith Winston @SmithWinst13560
6 Followers 742 Following Engineer; Researcher; Working in CV, ML and MLSys
Scott Santens @scottsantens
145K Followers 133K Following Founder and CEO of @ITSAfoundation; I focus on unconditional universal basic income (UBI) | #BasicIncome #UBI Bluesky: https://t.co/pPlwJvh31P
Astitva Srivastava @sarcastitva
536 Followers 382 Following Research Scientist (Intern) @ Meta | Google PhD Fellow | CVIT @ IIITH
Siddharth Katageri @siddKatageri
104 Followers 234 Following Currently MS by Research @iiit_hyderabad. Working on 3D Vision, Human-Scene Interaction.
Varun Gupta @__varungupta__
44 Followers 423 Following Teaching Machines How to See 🎬 🎓 MS-CSE @iiit_hyderabad | Center for Visual Information Technology (CVIT)
Yu Sun @YuSunMark
836 Followers 378 Following Assistant Professor @JHUECE, Postdoc @Caltech, Ph.D. @wustlcig, Researcher in Computational imaging.
AK @_akhaliq
428K Followers 3K Following AI research paper tweets, ML @Gradio (acq. by @HuggingFace 🤗) dm for promo ,submit papers here: https://t.co/UzmYN5YmrQ
Yann LeCun @ylecun
955K Followers 765 Following Professor at NYU. Chief AI Scientist at Meta. Researcher in AI, Machine Learning, Robotics, etc. ACM Turing Award Laureate.
Michael Black @Michael_J_Black
85K Followers 706 Following Director, Max Planck Institute for Intelligent Systems (@MPI_IS). Chief Scientist @meshcapade. Building 3D digital humans using vision, graphics, and learning.
Andrej Karpathy @karpathy
1.4M Followers 1K Following Building @EurekaLabsAI. Previously Director of AI @ Tesla, founding team @ OpenAI, CS231n/PhD @ Stanford. I like to train large deep neural nets.
Sebastian Raschka @rasbt
358K Followers 1K Following ML/AI researcher & former stats professor turned LLM research engineer. Author of "Build a Large Language Model From Scratch" (https://t.co/O8LAAMRzzW).
Kosta Derpanis @CSProfKGD
69K Followers 197 Following #CS Assoc Prof @YorkUniversity, #ComputerVision Scientist Samsung #AI, @VectorInst Faculty Affiliate, TPAMI AE, @ELLISforEurope Member #ICCV2025 Publicity Chair
Peyman Milanfar @docmilanfar
94K Followers 501 Following Distinguished Scientist at Google. Computational Imaging, Machine Learning, and Vision. Tweets = personal opinions. May change or disappear over time.
Danish Pruthi @danish037
11K Followers 707 Following Faculty at the Indian Institute of Science, Bangalore. PhD from @LTIatCMU.
Jia-Bin Huang @jbhuang0604
65K Followers 283 Following
F. Güney @ftm_guney
9K Followers 2K Following research on computer vision, teaching, and movies. tweets in TR, EN
Dima Damen @dimadamen
10K Followers 699 Following Professor of Computer Vision, @BristolUniEng. Senior Research Scientist @GoogleDeepMind - passionate about the temporal stream in our lives.
Alfredo Canziani @alfcnz
119K Followers 296 Following Musician, math lover, cook, dancer, 🏳️🌈, and an ass prof of Computer Science at New York University
#CVPR2026 @CVPR
49K Followers 330 Following Official account for IEEE/CVF Conference on Computer Vision & Pattern Recognition. Hosted by @CSProfKGD with more to come.
Ishan Misra @imisra_
7K Followers 245 Following Director, GenAI Research @Meta. Tech Lead of Movie Gen Meta. Past: MIT TR35, Llama3, Emu Video, ImageBind, DINO. Tweets and opinions are my own.
Makarand Tapaswi @MakarandTapaswi
1K Followers 632 Following Principal ML Scientist @WadhwaniAI | Assistant Professor @IIIT_Hyderabad | Opinions my own
Weights & Biases @weights_biases
46K Followers 1K Following The AI developer platform.🛠️ Track and evaluate your LLM applications in real-time with @weave_wb.
Richard Socher @RichardSocher
113K Followers 1K Following CEO @youdotcom MP @aixventuresHQ Before: Stanford Adj Prof in AI/NLP, Chief Scientist at Salesforce, MetaMind
Jiasen Lu @jiasenlu
1K Followers 370 Following Researcher@Apple (Ex Senior researcher @AI2, PhD@Georgia Tech)
Noam Razin @noamrazin
670 Followers 333 Following Postdoctoral Fellow at @PrincetonPLI | Past: Computer Science PhD @TelAvivUni & Apple Scholar in AI/ML | Interested in the foundations of deep learning
Akari Asai @AkariAsai
19K Followers 870 Following Incoming Assistant Professor @SCSatCMU & research scientist @allen_ai. akariasai @ 🦋
Tsinghua KEG (THUDM) @thukeg
4K Followers 170 Following #ChatGLM #GLM130B #CodeGeeX #CogVLM #CogView #AMiner The Knowledge Engineering Group (KEG) and THUDM at @Tsinghua_Uni @jietang @ericdongyx
slime @slime_framework
208 Followers 3 Following The LLM post-training framework for RL Scaling. https://t.co/4ILpx8hfKN
Wenting Zhao @wzhao_nlp
5K Followers 608 Following reasoning & llms @Alibaba_Qwen Opinions are my own
Vaish Shrivastava @VaishShrivas
364 Followers 420 Following Reinforcement Learning @MSFTResearch 🧠 MS CS @Stanford @StanfordNLP @StanfordAILab BS CS @Caltech
Google AI Studio @GoogleAIStudio
53K Followers 2 Following The fastest path from prompt to production with Gemini
Scott Jeen @enjeeneer
689 Followers 2K Following Predicting the future @_Mantic_AI. Previously PhD at Cambridge University. AI and reinforcement learning.
Minhyuk Sung @MinhyukSung
2K Followers 635 Following Associate professor @ KAIST | KAIST Visual AI Group: https://t.co/mblvQKFc8t.
Ai2 @allen_ai
74K Followers 410 Following Breakthrough AI to solve the world's biggest problems. › Join us: https://t.co/MjUpZpKPXJ › Newsletter: https://t.co/k9gGznstwj
Ross Taylor @rosstaylor90
10K Followers 1K Following Building @GenReasoning. Previously lots of other things like: Llama 3/2, Galactica, Papers with Code.
Qianqian Wang @QianqianWang5
3K Followers 439 Following Incoming Assistant Professor at Harvard and Kempner Institute. Postdoc at UC Berkeley. Former Ph.D. student at Cornell Tech. https://t.co/LyIdb5HmM9
Sathwik @VishnuSathvik1
770 Followers 705 Following Trying to make machines reason @precogatiiith | ug @iiit_hyderabad | Tweets everything.
Mehul Damani @MehulDamani2
598 Followers 405 Following PhD-ing @mit @MIT_CSAIL | language models, reinforcement learning
Roberta Raileanu @robertarail
9K Followers 2K Following Senior Staff Research Scientist @GoogleDeepMind & Honorary Lecturer @UCL. ex @Meta|@MSFTResearch|@NYU|@Princeton. Llama-3, Toolformer, Rainbow Teaming, MLGym.
Pascale Fung @pascalefung
3K Followers 46 Following Senior Director of A.I. Research, Meta-FAIR. Chair Professor of ECE, HKUST. Fellow of AAAI, ACL, IEEE, ISCA.
Arnaud Doucet @ArnaudDoucet1
6K Followers 811 Following Senior Staff Research Scientist @GoogleDeepMind, previously Oxford Stats Prof - interested in Comp Stats, Generative Modeling, Monte Carlo, Optimal Transport
Guan-Horng Liu @guanhorng_liu
1K Followers 370 Following Research Scientist @MetaAI (FAIR NY) • Schrödinger Bridge, diffusion, flow, stochastic optimal control • prev ML PhD @GeorgiaTech 🚀
Balaji Krishnamurthy @KbalajiTweets
512 Followers 278 Following Senior Principal Scientist and Senior Director, Media and Data Science Research Lab, Adobe
Mustafa Suleyman @mustafasuleyman
170K Followers 486 Following CEO, Microsoft AI | Author: The Coming Wave | Past: Co-founder, @InflectionAI & @GoogleDeepMind
Songlin Yang @SonglinYang4
14K Followers 3K Following research @MIT_CSAIL @thinkymachines. work on scalable and principled algorithms in #LLM and #MLSys. in open-sourcing I trust 🐳. she/her/hers
Amath SOW @AmathSo36561819
776 Followers 3K Following PhD student at Reasoning and Learning Lab(Real), Linkoping(Sweden): Multi-agent planning in dynamic and uncertain env, RL , meta heuristics, Robotics
Aadith Warrier @AadithWarrier
21 Followers 341 Following Robotics and Vision | Research @ RRC IIIT-H | MechE @ BITS Pilani '25
Armen Aghajanyan @ArmenAgha
15K Followers 285 Following Co-founder & CEO @perceptroninc; ex-RS FAIR/MSFT
Zi Wang, Ph.D. @ziwphd
1K Followers 183 Following Research Scientist @ Google DeepMind. CS PhD @ MIT CSAIL. Opinions my own. https://t.co/UJRqylN2Df
Natalie Schluter @natschluter
5K Followers 470 Following #NoJusticeNoPeace-- Machine Learning Researcher at Apple MLR-- All tweets/opinions my own
Jules @julesagent
22K Followers 9 Following Your new async coding agent by @GoogleLabs. Built for devs, open to feedback, evolving with you. Dive in → https://t.co/iIzFEMmWgv
Michele Caprio @mic_caprio
482 Followers 667 Following Lecturer (Asst. Prof.) @csmcr. On this profile, math (Imprecise Probabilities/Stats/AI/ML), baseball (@Phillies), and football (@juventusfcen).
Zan Gojcic @ZGojcic
3K Followers 645 Following Research manager at @NVIDIAAI working on neural reconstruction and data-driven simulation.
Harkirat Behl @BehlHarkirat
2K Followers 322 Following Crafting Phi models @ Microsoft Research ''Synthetic Data''
Prathyush Sambaturu @prathyushspeaks
2K Followers 515 Following Postdoc @UniofOxford @OxfordBiology @PSIOxford | Ph.D. @uva | MS @iiit_hyderabad | he/him
James Zou @james_y_zou
16K Followers 119 Following @Stanford professor. Chan-Zuckerberg investigator. Sloan Fellow. Overton Prize. @togethercompute. AI for science + health.
Danqi Chen @danqi_chen
17K Followers 789 Following Associate professor @princetonPLI @PrincetonCS. Previously: @facebookai, @stanfordnlp, @Tsinghua_Uni https://t.co/hLXgShMIPE
Unsloth AI @UnslothAI
32K Followers 459 Following Open source LLM fine-tuning & RL! 🦥 https://t.co/2kXqhhvLsb
Vincent Sitzmann @vincesitzmann
16K Followers 307 Following Teaching AI to see, model, and interact with our 3D world. Assistant Professor @ MIT, leading the Scene Representation Group (https://t.co/h5gvhLYrtw).
Jon Barron @jon_barron
30K Followers 1K Following AI researcher at Google DeepMind. Synthesized views are my own.
Aslan Pahari @AslanPahariii
21K Followers 158 Following Views are my own. Follow ≠ Endorsement. Usually talking about anthropology, history, linguistics and other related stuff.
Isha Puri @ishapuri101
803 Followers 418 Following AI / NLP PhD-ing @MIT_CSAIL @nlp_mit, currently product @AbridgeHQ prev @Harvard /HBS
Arpita Biswas @arpitabiswas777
1K Followers 163 Following Assistant professor @RutgersCS • Research Associate @HSPH • @HCRCS Postdoctoral Fellow@Harvard • Google Ph.D. Fellow@IIScCSA • @GoogleResearch• @MSFTResearchTrends for United States
You might like
