Image created with gemini-3.1-flash-image-preview with claude-sonnet-4-5. Image prompt: Photorealistic 4K nature shot of a winter bay at dusk with deep blue to orange sunset gradient, a large rectangular portal carved into thick ice sheet showing tropical beach scene within, ice crystals framing edges, dark water around ice, golden hour lighting catching ice textures, National Geographic quality, landscape format with bold sans-serif AR/VR title text.
Runway News | New Funding to Scale World Simulation https://runwayml.com/news/runway-series-e-funding
The Waymo World Model: A New Frontier For Autonomous Driving Simulation https://waymo.com/blog/2026/02/the-waymo-world-model-a-new-frontier-for-autonomous-driving-simulation/
This model is incredibly impressive and a massive step forward for autonomous driving!! 🚘 Huge congrats to the @Waymo team, and a special shoutout to the project’s main driver @maxjiang93! Thanks for having me to have contributed a small piece to this one🥂”” https://x.com/songyoupeng/status/2019828959660372387
We’re excited to introduce the Waymo World Model–a frontier generative mode for large-scale, hyper-realistic autonomous driving simulation built on @GoogleDeepMind’s Genie 3. By simulating the “impossible”, we proactively prepare the Waymo Driver for some of the most rare and”” https://x.com/Waymo/status/2019804616746029508
High speed FPV drones got the winter olympics looking like a real life video game. Really makes you appreciate whoever wrote the rendering stack for reality.”” https://x.com/bilawalsidhu/status/2021207946240078271
This will change the way we experience sports forever — watching the game from a gods eye view. Arcturus is building 4D gaussian splatting tech that can capture every angle of a sporting event and pushes the bar for volumetric video. I tested this in a headset and it makes 360″” https://x.com/bilawalsidhu/status/2019831831110258883
[2602.05842] Reinforcement World Model Learning for LLM-based Agents https://arxiv.org/abs/2602.05842
SceneLinker: Compositional 3D Scene Generation via Semantic Scene Graph”” TL;DR: a 3D scene synthesis approach that predicts semantic scene graphs from RGB sequences and generates layout-consistent 3D environments”” https://x.com/Almorgand/status/2020868819531546800
The tech we use to capture sports is a decade ahead of how we actually watch it. We can reproduce entire games in 3D, track every player down to their pose and heartbeat – but barely any of that makes it to your living room. 0:00 Immersion Gap 0:48 Computer Vision & Hawk-Eye”” https://x.com/bilawalsidhu/status/2020229779585200401
Clean workflow to reskin reality 360 pano > nano banana > world labs Big limit is the navigable volume. Once we can auto fuse multiple 360 panos this approach gets v compelling. Y capture so much 3d data when are you need are a few nicely sampled panos?”” https://x.com/bilawalsidhu/status/2021702814376894653
PixVerse R1 is here! Unlock the future of real-time interactive worlds in 720P. What’s more? Experience endless possibilities for creation in R1 community. RT+Reply+Follow=300Creds (72H ONLY) + Invite Code (500 RANDOM WINNERS) in DMs”” https://x.com/PixVerse_/status/2021486175391973489
Humans in VR headsets piloting humanoid robots in a cage match. I love that this is your Saturday in San Francisco in 2026. The homies at REK have an amazing event this Saturday in SF – the night before the Super Bowl. Code BILAWAL gets you 50% off tickets. Not sponsored; just”” https://x.com/bilawalsidhu/status/2019545022073598233
Joel Jang https://joeljang.github.io/world-models-for-robotics
Not the flashiest demos, but what’s under the hood represents a foundational shift for general-purpose robotics. World models are the next-gen foundation of Physical AI, not the VLM backbones found in typical VLAs. DreamZero is a 14B-parameter World Action Model (WAM) by NVIDIA”” https://x.com/TheHumanoidHub/status/2019460701811851593
Robots usually fail for one simple reason: they don’t understand what will happen next. [📍Paper, code & task gallery at the end] > 14B “World Action Model” from @nvidia: DreamZero… Instead of copying motions or replaying demonstrations, this model predicts how the world”” https://x.com/IlirAliu_/status/2019418751976800520
Leave a Reply