Top Tweets for #worldModels
Will be attending @RL_Conference!
Presenting our work on **Temporally-extended Latent Action World Models** at the #RLC2026 MBRL+WM Workshop đ
Looking forward to hearing perspectives on the role of world models in sequential decision-making
#WorldModels #MBRL #HRL

AI is moving from generating scenes to generating worlds.
NVIDIA Lyra 2.0 turns walkthroughs into persistent, explorable 3D environments with spatial continuity.
Full NexusAI analysis:
https://t.co/BPZv9MNCpR
#NVIDIA #SpatialAI #WorldModels #Generative3D #NexusAI
https://t.co/EETMG05Ea1 â World Action Models Ă AGI.
A name built for the next generation of AI: models that donât just understand the world â they predict, plan, and act within it.
WAMAGI â Powering the path to AGI.
#AI #AGI #WorldModels #WAM #Robotics

¿Qué busca Anthropic con la compra de Decart AI por 6.000 millones? #Anthropic #DecartAI #ClaudeAI #InteligenciaArtificial #IA #ArtificialIntelligence #Nvidia #Tecnologia #Startups #BigTech #WorldModels #Claude #felizjueves #13deagosto https://t.co/MJdPtMvxMV
AI data centers are straining power grids and drawing scrutiny. @mpvizard @ashimel @jonswartz @AndiMann @terinrobinson break down the energy costs, the rise of world models, and the Silk Typhoon takedown. #AI #DataCenters #Cybersecurity #WorldModels https://t.co/qHzdnUEI2F
Work with my co-authors Gaytri Jena
Aman Chadha , Vinija Jain, Vasu Sharma, Amitava
Questions and feedback very welcome đ
#MachineLearning #ComputerVision #WorldModels
World models : la prochaine rĂ©volution de lâIA ?
https://t.co/at5iYyIvia
#worldModels #science #apprentissageParRenforcement #robotique #deepLearning #intelligenceArtificielle
World models are often trained for short steps, then recursively rolled out much further. DPWM instead predicts a long-horizon endpoint directly, and gains grew with horizon length.
Source: https://t.co/LwPBm4i14n
#WorldModels #AIResearch
Follow for daily AI research.
The World Model and Spatial Intelligence Era: Governing AI Beyond Language #AI #Worldmodels https://t.co/vh8WG3xwwG
Just updated the Humanist Neuro Symbolic World Model! đ§ This architecture builds on Yann LeCun's concepts, utilizing a JEPA encoder to ground perception
https://t.co/CE22qxom5b
#NeuroSymbolic #AI #WorldModels #JEPA #LeCun #MachineLearning#PatientAdvocate

Leap out of the box, unleash the final artifact, open-sourced to the world.
#AgenticAI #QuantumComputing #Robotics #OpenWeights #DeepTech #AutonomousSystems #WorldModels #EdgeAI #NextGenTech #Innovation #FutureOfTech #SoftwareEngineering #Invention #AIDev #SpatialComputing
Excited to share that I'm a Winner of the Qualcomm Innovation Fellowship 2026! đ
My proposal, "Towards Scene State Tokenization for World Models" focuses on 4D representation learning for efficient generative world modeling.
https://t.co/xRbP9VLL6i
#Qualcomm #AI #WorldModels
Fei-Fei Li. Jeff Dean. Terence Tao. Steven Chu. And more.
The AASF Summit brings the world's top Scientists, Nobel laureates & a Fields medalist to Stanford, Aug 7â9.
đïž https://t.co/o41CYWkfdd · 10% off with aasf-cui-10off
#AI #WorldModels #AASF
Built a voice driven ghost hunting adventure in an abandoned asylum, powered by @reactorworld, a @runwayml character/avatar as a guide and world model Happy Oyster!
First results here!
#worldmodels #happyoyster
Robotics needs models that can predict the world and act within it. WorldDiT pushing that frontier without a VLM generating each action is a meaningful step toward faster, more deployable robot intelligence.
#Robotics #WorldModels
We are releasing WorldDiT, a unified architecture for robotics world modeling and control.
On the LIBERO benchmark, it performs the best among all publicly released methods that do not need a VLM to generate actions. Its size and performance sit on the reported Pareto frontier.

just had the best time at @siggraph this year! so good seeing old friends again and meeting new ones too. excited for where this industry is headed and even more excited for what we're building next âș
#SIGGRAPH2026 #WorldModels #ComputerScience

La convergence physique de lâIA gĂ©nĂ©rative đ€
Et si la vidĂ©o gĂ©nĂ©rative nâĂ©tait plus seulement faite pour nos yeux, mais pour piloter le monde rĂ©el ?
#KingLand #IA #Robotique #FLUX3 #BlackForestLabs #WorldModels #PhysicalAI #TechVision #Innovation #DeepTech #VideoIA #Multimodal
â«ïž Fiche Impact : https://t.co/a2sNRSbYkZ
Parce quâavec FLUX 3, un modĂšle physique unifiĂ© vaut mieux quâune mosaĂŻque dâoutils dĂ©connectĂ©s.
La gĂ©nĂ©ration vidĂ©o sâest longtemps limitĂ©e Ă une projection dâimages esthĂ©tiques mais creuses, sans aucune notion de gravitĂ©, dâacoustique ou de collision. @bfl_ai Black Forest Labs brise ce plafond de verre avec FLUX 3. En unifiant lâimage, la vidĂ©o, le son synchrone et lâaction mĂ©canique au sein dâun seul espace latent, ce modĂšle de monde transforme lâIA gĂ©nĂ©rative en un vĂ©ritable simulateur de physique gĂ©nĂ©rale prĂȘt pour lâindustrie.
đŠ Une architecture Omni unique qui fusionne les pixels, les pistes audio et les trajectoires de commande sans aucune latence de traduction.
â±ïž Une synchronisation sensorielle parfaite capable de gĂ©nĂ©rer 20 secondes de vidĂ©o avec des effets sonores calĂ©s Ă la milliseconde prĂšs.
đŠŸ Le dĂ©codeur FLUX-mimic qui permet aux bras robotiques dâapprendre la manipulation fine directement par imitation visuelle.
đ Un impact opĂ©rationnel immĂ©diat avec une rĂ©duction de 70% du temps dâintĂ©gration des pipelines de production.
"Nous avons longtemps traitĂ© la vidĂ©o gĂ©nĂ©rative comme une simple toile animĂ©e pour divertir les esprits. En y intĂ©grant les lois de la physique et la motricitĂ©, nous passons de lâĂšre de lâIA spectatrice Ă celle de lâIA actrice, capable de comprendre son environnement avant mĂȘme de le toucher."
â C. Pestel
En observant cette convergence, jâĂ©prouve une fascination mĂȘlĂ©e dâun profond rĂ©alignement de mes certitudes. On a souvent pensĂ© que la robotique et la crĂ©ation artistique appartenaient Ă des mondes Ă©tanches. Pourtant, voir un mĂȘme modĂšle mathĂ©matique synchroniser un bruit de verre avec une prĂ©cision millimĂ©trique et guider un bras mĂ©canique pour manipuler un objet souple change la donne. Cette plasticitĂ© extrĂȘme des modĂšles de monde nous force Ă repenser nos mĂ©tiers : demain, concevoir une campagne vidĂ©o et programmer une chaĂźne logistique industrielle feront appel Ă la mĂȘme intelligence sous-jacente.
đ Fiche Tool : https://t.co/h6S73ZIG7J
Comment imaginez-vous collaborer avec une IA qui comprend la physique de notre monde aussi bien que nous ?

Un bon article d'introduction à JEPA (Joint-Embedding Predictive Architecture), famille d'architectures IA appartenant au domaine des World Models pronée par Yann LeCun #IA #WorldModels https://t.co/6NhW0pxHzI
Last Seen Hashtags on Sotwe
omegle() or #ometv() or #teenageer() or #monkey()+filter:videos
Seen from Brazil
èȘćœ
Seen from Switzerland
momsonn
Seen from United States
àčàžàčàž«àžàčàžąàžČàž§
Seen from Thailand
ebony femboy
Seen from United States
怫怫
Seen from Turkey
æèŠéćć©
Seen from United States
str8 baited
Seen from Turkey
irishchav
Seen from United Kingdom
ÙÙŰŹ_Űč۱ۧÙÙ
Trends for you
Most Popular Users

Elon Musk 
@elonmusk
241.3M followers

Barack Obama 
@barackobama
119.1M followers

Cristiano Ronaldo 
@cristiano
113.1M followers

Donald J. Trump 
@realdonaldtrump
111.8M followers

Narendra Modi 
@narendramodi
107.1M followers

Rihanna 
@rihanna
98.4M followers

NASA 
@nasa
92.3M followers

Justin Bieber 
@justinbieber
91.5M followers

KATY PERRY 
@katyperry
89.2M followers

Taylor Swift 
@taylorswift13
83.1M followers

Lady Gaga 
@ladygaga
74.6M followers

Virat Kohli 
@imvkohli
72.2M followers

Kim Kardashian 
@kimkardashian
70.5M followers

YouTube 
@youtube
68.8M followers

Neymar Jr 
@neymarjr
65.1M followers

Bill Gates 
@billgates
64.7M followers

The Ellen Show
@theellenshow
62.4M followers

Selena Gomez 
@selenagomez
62.3M followers

CNN 
@cnn
61.8M followers

X 
@x
60.8M followers























