From Generation to Simulation: How Far Are World Models from Being True Simulators?

Hugging Face Daily Papers Papers
world-models generative-ai simulation ai-research video-generation physical-simulation capability-assessment

Summary

This paper systematically evaluates generative world models against traditional simulators across eight capabilities, finding progress in interaction and controllability but gaps in physical guarantees, state feedback, and long-horizon stability.

With the rapid progress of diffusion models and large-scale video generation, generative world models are increasingly expected to replace traditional simulators, including physics engines, game engines, and reinforcement-learning environments. Yet the remaining distance from generation to simulation lacks a systematic assessment. We present a capability-based study using an external yardstick: eight capabilities of a traditional simulator, namely asset construction, physics engine, interaction, controllability, stability, state feedback, diversity, and evaluation metrics. We trace three main technical routes--latent dynamics, video generation, and joint-embedding prediction--and map exactly 200 representative works published from 2018 to June 2026 onto these capabilities. Our analysis shows that world models have achieved functional substitution in interaction and controllability for specific scenarios, but remain short of traditional simulators in formal guarantees of physical laws, structured state feedback, and reproducible long-horizon evolution. State feedback is the most neglected cross-route shortcoming: only 6 of 163 implementation papers expose a runtime interface for querying entity states or physical parameters. We identify six research directions: formalized physics, a unified action interface, first-class state feedback, long-horizon stability, downstream-utility evaluation, and cross-route hybridization. Project page: https://github.com/AtongWang/world-model-simulators
Original Article
View Cached Full Text

Cached at: 08/25/26, 08:34 AM

Paper page - From Generation to Simulation: How Far Are World Models from Being True Simulators?

Source: https://huggingface.co/papers/2608.23070

Abstract

Generative world models are evaluated against traditional simulators across eight capabilities, revealing gaps in physical guarantees, state feedback, and long-horizon stability despite progress in interaction and controllability.

With the rapid progress ofdiffusion modelsand large-scalevideo generation,generative world modelsare increasingly expected to replace traditional simulators, includingphysics engines, game engines, and reinforcement-learning environments. Yet the remaining distance from generation to simulation lacks a systematic assessment. We present a capability-based study using an external yardstick: eight capabilities of a traditional simulator, namely asset construction,physics engine, interaction,controllability, stability,state feedback, diversity, and evaluation metrics. We trace three main technical routes--latent dynamics,video generation, andjoint-embedding prediction--and map exactly 200 representative works published from 2018 to June 2026 onto these capabilities. Our analysis shows that world models have achieved functional substitution in interaction andcontrollabilityfor specific scenarios, but remain short of traditional simulators in formal guarantees of physical laws, structuredstate feedback, and reproducible long-horizon evolution.State feedbackis the most neglected cross-route shortcoming: only 6 of 163 implementation papers expose a runtime interface for querying entity states or physical parameters. We identify six research directions: formalized physics, a unified action interface, first-classstate feedback,long-horizon stability, downstream-utility evaluation, andcross-route hybridization. Project page: https://github.com/AtongWang/world-model-simulators

View arXiv pageView PDFProject pageGitHubAdd to collection

Get this paper in your agent:

hf papers read 2608\.23070

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/2608.23070 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.23070 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.23070 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

Coding Agent Is Good As World Simulator

arXiv cs.AI

This paper presents an agentic framework that uses coding agents to generate physically plausible world simulations from natural language prompts, outperforming video-based models in physical accuracy and instruction fidelity.

PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives

Hugging Face Daily Papers

PlayWorld is a benchmark for evaluating interactive video world models using multi-modal agent players pursuing long-horizon objectives. It assesses geometry consistency, interaction fidelity, and state evolution, revealing that current models struggle with spatial consistency and persistent state evolution.

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.