Stanford's T2-z lets robots probe hidden physics with a poke, lifting success from 34% to 73%
stanfordnlp · x · 2026-10-21
Stanford's T2-z tackles the hidden physics (slipperiness, friction, mass) that breaks robots in new environments.
- A quick test-time "poke" tunes only a tiny 32-number environment code; the world model weights stay frozen (even 1 sample works).
- On real robots: ball friction tasks went 0% → 62.5%, stick balance 20% → 83%.
- Best on all 11 tasks against 4 baselines.
The takeaway: a quick probe beats retraining when unseen physics is the failure mode.
More from Embodied
- NVIDIA open-sources MotionBricks: 350K motion skills at 15,000 FPS, 2ms latency — aftahi_ai · 2026-10-21
- FLoTS-SLAM uses optical-flow residuals to hit 0.73 DM-IoU in dynamic monocular SLAM — zhenjun_zhao · 2026-10-21
- Plato ring launches at $199 to turn near-silent speech into AI commands — TinfoilTricorn · 2026-10-21
- Dream-1 controlled 31 unseen robot types across 53 countries within 24 hours — gharik · 2026-10-21
- Snap Hires Robert Downey Jr. as Chief Customer Experience Officer to Sell Specs AR Glasses Against Meta — adariostrange · 2026-10-21
- Fei-Fei Li team unveils FACT, a force-aware foundation model for robot dexterity — stanfordnlp · 2026-10-21