Artificial Intelligence · 24.08.2026, 07:16 UTC
What a World Model Represents Is Three Questions
| Schweregrad | info |
|---|---|
| Kategorie | Artificial Intelligence |
| Quelle | arXiv cs.AI ↗ |
| Veröffentlicht | 24.08.2026 UTC |
Sicherheitsmeldung mit Schweregrad noch nicht bewertet. Technische Details im Tab „Originaltext“; empfohlene Schritte in der Checkliste.
arXiv:2607.06640v2 Announce Type: replace-cross Abstract: World models learn task-relevant information through many routes: observation reconstruction, recurrent state, temporal filtering, and explicit task supervision. Different routes can make different variables available. The same variable can also be available through several routes at once. When it is, looking at which route would increase the training loss most if removed does not tell you which route the model actually uses. The questions are reachability, whether a training signal can identify a task-relevant direction; admission, whether that direction is recoverable from the latent; and assignment, which eligible route carries it. We test them in environments with a known set of required coordinates. A direction cannot enter the latent unless some training signal can identify it. Reconstruction, recurrence, or filtering may already recover some of those coordinates; a reward or value head then has no residual direction to admit. For what remains, how many independent predictions the target supplies is how many coordinates install: one through four independent predictions admit one through four directions, including through the value head. Reachability is not admission: a temporal second-moment coefficient can remain absent under next-token prediction when accumulating it is a fraction of a percent of that loss, and a head that predicts the coefficient restores it. Assignment is a different test. Two routes that each carry the same variable when trained alone do not swap the carrier when we reverse which is more …
Maßnahmen
⬇ Als MarkdownVerwandte Beiträge
- info COMET: Contrastive Motion-Enhanced Temporal Reasoning for Video Multimodal Large Language Models
- info From a Static Multi-Level Small Semantic Codebook to a Dynamic Single-Level Large Semantic Codebook for Generative Recommendation
- info EviRank: Structured Relevance Evidence for Multimodal Image Re-ranking
- info Sharing the Control Authority Between Deep Reinforcement Learning and Model Predictive Control: Application to Multi-Class Transportation Networks