robotics
One Trial Teaches the World Model What the Camera Cannot See
T2-Z adapts a compact environment latent at test time to represent friction, mass and other hidden conditions.
Summary
T2-Z adapts a compact environment latent at test time to represent friction, mass and other hidden conditions.
A world model can observe an object without seeing the friction, mass or local rule that governs what happens next. T2-Z keeps the model frozen at deployment and optimizes a newly initialized environment latent from one or more interactions. Across seven simulated settings and four real robot tasks, the authors report object-prediction error falling from 15.20 to 7.00 and action success rising from 33.73 to 72.96 percent. Some useful latent structure emerged from a single trajectory. The evaluation does not establish that the learned coordinates correspond cleanly to every physical property, but it offers a compact route to test-time adaptation.
Why it matters
T2-Z adapts a compact environment latent at test time to represent friction, mass and other hidden conditions.
Limits and context
- The evaluation does not establish that the learned coordinates correspond cleanly to every physical property, but it offers a compact route to test-time adaptation.
Key claims
T2-Z adapts a compact environment latent at test time to represent friction, mass and other hidden conditions.
Qualification: The evaluation does not establish that the learned coordinates correspond cleanly to every physical property, but it offers a compact route to test-time adaptation.
Evidence: source-2026-10-17-004
Sources
- arXiv preprint 2610.22364, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.