Introduces looped transformer architectures for world models that iteratively refine latent states to achieve up to 100x parameter efficiency via adaptive computation depth.
Ruoyao Wang, Peter Jansen, Marc-Alexandre C ˆot´e, and Prithviraj Ammanabrolu
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Looped World Models
Introduces looped transformer architectures for world models that iteratively refine latent states to achieve up to 100x parameter efficiency via adaptive computation depth.