🛰️ Daily AI Frontier
‹ back to 2026-08-25

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

Research Multimodal & Generative

Ranking

Overall 80
Content 90
Popularity 57

Observed public metrics from 1 member.

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

Merged summary

TL;DR - This survey evaluates whether generative world models can replace traditional simulators by mapping 200 works across eight core simulation capabilities. Current models can support interaction and controllability in specific settings, but still lack reliable physics, structured state feedback, and reproducible long-horizon behavior.

  • Compares three technical routes: latent dynamics, video generation, and joint-embedding prediction.
  • Uses traditional simulators as an external benchmark across asset construction, physics, interaction, controllability, stability, state feedback, diversity, and evaluation.
  • State feedback is the largest cross-route gap: only 6 of 163 implementation papers provide runtime access to entity states or physical parameters.
  • Priorities include formalized physics, unified action interfaces, first-class state feedback, long-horizon stability, utility-based evaluation, and hybrid approaches.

Sources (1)

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

arXiv cs.AI Tong Wang, Huan Deng, Mucheng Yang, Yang He, Xiaohui Kuang, Gang Zhao 2026-08-24 arXiv:2608.23070
Public signals Hugging Face upvotes 4
Providers: Hugging Face · Upvotes 4 OpenAlex · N/A Publisher · N/A Semantic Scholar · N/A X · N/A Fetched 2026-09-24 14:34:09.184580 UTC

TL;DR - This survey evaluates whether generative world models can replace traditional simulators by mapping 200 works across eight core simulation capabilities. Current models can support interaction and controllability in specific settings, but still lack reliable physics, structured state feedback, and reproducible long-horizon behavior.

  • Compares three technical routes: latent dynamics, video generation, and joint-embedding prediction.
  • Uses traditional simulators as an external benchmark across asset construction, physics, interaction, controllability, stability, state feedback, diversity, and evaluation.
  • State feedback is the largest cross-route gap: only 6 of 163 implementation papers provide runtime access to entity states or physical parameters.
  • Priorities include formalized physics, unified action interfaces, first-class state feedback, long-horizon stability, utility-based evaluation, and hybrid approaches.
item →