/1 Meet LoopWM: The first looped transformer for stable, long-horizon world modeling! Current world models struggle with deep computation costs and compounding errors. Enter Looped World Models (LoopWM), which iteratively refine latent states through a parameter-shared
ROBOTICS
-
Physical AI: Leading Companies Bringing AI into Robots and Machines
By
–
#PhysicalAI is moving artificial intelligence from screens and software into #robots, #drones, self-driving #vehicles and #intelligent #machines. Here are the companies #leading the race to bring #AI into the real world and why their work could reshape #business, work and
-
Software updates reduce errors, earning trust for long autonomous drives.
By
–
It gets rid of mistakes with every update. Not many left. It has been earning more and more trust. I drove from Seattle to San Jose in one shot two years ago. Only rarely had my hand on the steering wheel. Even through smoke, fog, snow, and heavy rain.
-
Learning world models from dropping a cup of water
By
–
When @olivercameron was the first to teach us about world models (he just collected $300 million investment last week) in my head I was thinking: "If I drop a cup on the ground, with some water in it, and film that with a high speed camera, the world model would learn a lot about
-
AI Robotics Challenges in 2030s Impact Employment and Pensions
By
–
Additional challenges will be posed by advanced AI robotics in the 2030s and their impact upon employment and pensions funding.
-
Who will finance pensions with AI robots?
By
–
The question remains: if robots equipped with advanced AI take over a large part of work in the 2030s, who will then fund pensions?
-
Top AI Papers of the Week: June 14-21 Highlights
By
–
The Top AI Papers of the Week (June 14 – June 21): – PreAct
– SpatialClaw
– Back on Track
– OpenClaw-Skill
– From Trainee to Trainer
– Compositional Skill Routing
– Can LLM Agents Infer World Models? Read on for more: -
Do as I Do: Dexerous Manipulation Data from Everyday Human Videos
By
–
“Do as I Do: Dexterous Manipulation Data from Everyday Human Videos” With how robot dexterity is bottlenecked by data as teleoperation and MoCap are expensive and internet videos are only observational, this paper turns normal RGB human videos into executable robot hand
-
T-Rex introduces basis hand motions for tactile learning
By
–
T-Rex introduces a new set of basis hand motions to learn tactile responses:
-
Offline agentic exploration develops robot skills
By
–
Great work using offline agentic exploration to develop robot skills! https://t.co/9JOJDpLohI
— Ken Goldberg (@Ken_Goldberg) 20 juin 2026Great work using offline agentic exploration to develop robot skills!