Rhoda study: web-video pre-training scales robot foundation models for real work
vincesitzmann · x · 2026-09-16
Robotics startup Rhoda published a study, Does Scaling Web-Video Pre-training Help Real Robots Do Real Work?, presenting what it calls the first evidence that robot foundation models can scale on internet video data.
- Based on thousands of trials and hundreds of hours of robot evaluations, the study shows scaling web-video pre-training leads to better real-world robot performance.
- Prior robot foundation models relied on teleoperation, UMI, or egocentric data; web video is data truly available anywhere.
- Amplifiers call this an exciting moment: scaling on internet video is a necessary requirement for massively powerful robot models.
More from Embodied
- Autonomous's Harness Pro uses an ESP32-P4, not a mini PC, to monitor AI coding agents — dee_hw · 2026-11-02
- KAIST's QCP lets robot policies control execution quality in real time — kaist-ai · 2026-11-02
- ANYbotics begins customer deliveries of ANYmal X, first legged robot certified for explosive atmospheres — lukas_m_ziegler · 2026-11-02
- InSpatio opens China's first embodied simulation data center targeting 1M hours/month — 智东西 · 2026-11-02
- GR-BC mines better trajectories from imperfect demos, hitting ~91% success on precise manipulation — 机器之心 · 2026-11-02
- RAM prices tripled, so Tesla cut a third of memory from Optimus's brain — aakashgupta · 2026-11-02