When driving and robot-learning papers say "world model", what exactly do they mean (inputs, outputs, intended use), and where does the word claim more than the evidence tests?
In every source inspected here a "world model" is a learned generative model that takes past observations (video) plus some control input and produces future video; the sources differ in what the model is for. GAIA-1 and Vista define it as predicting action consequences, NVIDIA as predicting the next observation given a "perturbation", 1X as an evaluator that also predicts task success. Several sources list uses (planning, policy training, evaluation) that they do not test; by 2026 Wayve and 1X frame the model as a closed-loop evaluation tool. The word is the authors' own label and none of the inspected definitions is tied to a physics-accuracy test.
In every source inspected here a "world model" is a learned generative model that takes past observations (video) plus some control input and produces future video; the sources differ in what the model is for. GAIA-1 and Vista define it as predicting action consequences, NVIDIA as predicting the next observation given a "perturbation", 1X as an evaluator that also predicts task success. Several sources list uses (planning, policy training, evaluation) that they do not test; by 2026 Wayve and 1X frame the model as a closed-loop evaluation tool. The word is the authors' own label and none of the inspected definitions is tied to a physics-accuracy test.
Curator qualifications: Sources agree a world model is a learned generator of future video from past observations plus control; they differ on purpose (action consequences, next state).
Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.
Structured fields
{
"citations": [
{
"url": "https://arxiv.org/abs/2309.17080v1",
"note": "candidate evidence excerpts"
},
{
"url": "https://arxiv.org/abs/2501.03575v3",
"note": "candidate evidence excerpts"
},
{
"url": "https://arxiv.org/abs/2405.17398v5",
"note": "candidate evidence excerpts"
},
{
"url": "https://arxiv.org/abs/2309.09777v2",
"note": "candidate evidence excerpts"
},
{
"url": "https://www.1x.tech/discover/1x-world-model",
"note": "candidate evidence excerpts"
},
{
"url": "https://www.1x.tech/1x-world-model.pdf",
"note": "candidate evidence excerpts"
},
{
"url": "https://wayve.ai/thinking/gaia-4/",
"note": "candidate evidence excerpts"
}
]
}Evidence claims
- [sourced_fact] In every source inspected here a "world model" is a learned generative model that takes past observations (video) plus some control input and produces future video; the sources differ in what the model is for. GAIA-1 and Vista define it as predicting action consequences, NVIDIA as predicting the next observation given a "perturbation", 1X as an evaluator that also predicts task success. Several sources list uses (planning, policy training, evaluation) that they do not test; by 2026 Wayve and 1X frame the model as a closed-loop evaluation tool. The word is the authors' own label and none of the inspected definitions is tied to a physics-accuracy test. · locator: candidate evidence excerpts source
Rights
| component | license | hosting | redistribution | modification | training | Evidence and notes |
|---|---|---|---|---|---|---|
| documentation | unknown | permitted | permitted | unknown | unknown | Applies to this original summary only, not to the cited sources. |
Linked records
- cites → https://arxiv.org/abs/2309.09777v2
- cites → https://arxiv.org/abs/2309.17080v1
- cites → https://arxiv.org/abs/2405.17398v5
- cites → https://arxiv.org/abs/2501.03575v3
- cites → https://wayve.ai/thinking/gaia-4
- cites → https://www.1x.tech/1x-world-model.pdf
- cites → https://www.1x.tech/discover/1x-world-model