Looped World Models

Hugging Face Daily Papers Papers

Summary

Looped World Models introduce iterative latent state refinement through shared transformer blocks, achieving 100x parameter efficiency while adapting computational depth to prediction complexity.

Current world models face a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this by introducing Looped World Models (LoopWM), which are the first looped architectures for world modelling. Our method iteratively refines latent environment states through a parameter-shared transformer block. This yield up to 100x parameter efficiency over conventional approaches with adaptive computation that automatically scales depth to match the complexity of each prediction step. Orthogonal to scaling model size and training data, LoopWM establishes iterative latent depth as a new scaling axis for world simulation, which might significantly push the community forward.
Original Article
View Cached Full Text

Cached at: 06/17/26, 03:35 AM

Paper page - Looped World Models

Source: https://huggingface.co/papers/2606.18208 Published on Jun 16

·

Submitted byhttps://huggingface.co/hongyyyyy

luon Jun 17

Authors:

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

,

Abstract

Looped World Models introduce iterative latent state refinement through shared transformer blocks, achieving 100x parameter efficiency while adapting computational depth to prediction complexity.

Currentworld modelsface a fundamental tension: faithful long-horizon simulation demands deep computation, but deeper models are expensive to deploy and prone to compounding errors. We resolve this by introducing LoopedWorld Models(LoopWM), which are the firstlooped architecturesfor world modelling. Our method iteratively refineslatent environment statesthrough aparameter-shared transformer block. This yield up to 100x parameter efficiency over conventional approaches withadaptive computationthat automatically scales depth to match the complexity of each prediction step. Orthogonal to scaling model size and training data, LoopWM establishes iterative latent depth as a new scaling axis for world simulation, which might significantly push the community forward.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2606\.18208

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2606.18208 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2606.18208 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2606.18208 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

World Machine: Towards Generative World Modeling for Time-Series

arXiv cs.LG

World Machine proposes a transformer-based generative world modeling architecture for time series that uses latent states to adapt to varying context lengths, addressing the quadratic memory cost of traditional transformers. Experiments on a synthetic dataset validate its feasibility and show improvements over conventional transformers.

@askalphaxiv: Another cool research on Looped Transformers They ask the question: "Can we loop a frozen, off-the-shelf checkpoint dir…

X AI KOLs Timeline

This research introduces a technique to loop frozen, off-the-shelf transformer checkpoints at inference time by using damped Runge-Kutta substeps, treating transformer layers as Euler steps in a residual ODE. This allows extra latent compute without fine-tuning, architecture changes, or new weights, showing gains on knowledge tasks like MMLU-Pro, GPQA, and ARC.

Next-Latent Prediction Transformers Learn Compact World Models

Papers with Code Trending

Introduces Next-Latent Prediction (NextLat), a self-supervised objective that trains transformers to predict their next latent state, encouraging compact internal world models and improving generalization across sequence modeling tasks.