AWhat changed
Learned a video tokenizer, latent-action representation and autoregressive dynamics from unlabeled videos to generate environments controllable frame by frame.
Scientific scope
Learned a video tokenizer, latent-action representation and autoregressive dynamics from unlabeled videos to generate environments controllable frame by frame. Limited learned virtual environments are not physically faithful simulators or unbounded coherent worlds. Text inputs may first require an external text-to-image system; absence of action labels is not absence of all training data.
Evidence
The linked original sources describe the stated contribution and its evidence limits. WRS has not independently rerun the experiments or certified a proof. Limited learned virtual environments are not physically faithful simulators or unbounded coherent worlds. Text inputs may first require an external text-to-image system; absence of action labels is not absence of all training data.
Contributors
- Source-listed authors and technical contributors · SCIENCE: Source-listed authors or explicitly credited technical contributors for the cited versions. Family entries retain source-specific attribution; a combined list is not one original-paper byline or equal individual work. Names alone do not establish resolved identities.
- Genie: Generative Interactive Environments
- Bruce et al.: Genie latent-action interactive environments — author/company project record
Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.
Full scientific record · Browse contribution grades · Request a correction