We survey world-model papers from the perspective of simulation rather than generation, asking a simple question: how far are current world models from becoming true simulators? We organize the field around eight simulator capabilities and identify key gaps in physics, state feedback, long-horizon stability, and evaluation. Hope this perspective can help connect world models with embodied AI, robotics, and interactive simulation. Project page: <a href=\"https://github.com/AtongWang/world-model-simulators\" rel=\"nofollow\">https://github.com/AtongWang/world-model-simulators</a></p>\n","updatedAt":"2026-08-25T08:28:28.362Z","author":{"_id":"665c65f1987f055eff6c8ea5","avatarUrl":"/avatars/980de0ae1964971259e051764a784c55.svg","fullname":"Frank W","name":"frankw132","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":1,"identifiedLanguage":{"language":"en","probability":0.8704923391342163},"editors":["frankw132"],"editorAvatarUrls":["/avatars/980de0ae1964971259e051764a784c55.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2608.23070","authors":[{"_id":"6a8cfd828dd056518b7f549b","user":{"_id":"665c65f1987f055eff6c8ea5","avatarUrl":"/avatars/980de0ae1964971259e051764a784c55.svg","isPro":false,"fullname":"Frank W","user":"frankw132","type":"user","name":"frankw132"},"name":"Tong Wang","status":"claimed_verified","statusLastChangedAt":"2026-08-25T08:14:07.448Z","hidden":false},{"_id":"6a8cfd828dd056518b7f549c","name":"Huan Deng","hidden":false},{"_id":"6a8cfd828dd056518b7f549d","name":"Mucheng Yang","hidden":false},{"_id":"6a8cfd828dd056518b7f549e","name":"Yang He","hidden":false},{"_id":"6a8cfd828dd056518b7f549f","name":"Xiaohui Kuang","hidden":false},{"_id":"6a8cfd828dd056518b7f54a0","name":"Gang Zhao","hidden":false}],"publishedAt":"2026-08-24T00:00:00.000Z","submittedOnDailyAt":"2026-08-25T00:00:00.000Z","title":"From Generation to Simulation: How Far Are World Models from Being True Simulators?","submittedOnDailyBy":{"_id":"665c65f1987f055eff6c8ea5","avatarUrl":"/avatars/980de0ae1964971259e051764a784c55.svg","isPro":false,"fullname":"Frank W","user":"frankw132","type":"user","name":"frankw132"},"summary":"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","upvotes":1,"discussionId":"6a8cfd828dd056518b7f54a1","projectPage":"https://github.com/AtongWang/world-model-simulators","githubRepo":"https://github.com/AtongWang/world-model-simulators","githubRepoAddedBy":"user","ai_summary":"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.","ai_keywords":["diffusion models","generative world models","latent dynamics","video generation","joint-embedding prediction","physics engine","controllability","state feedback","long-horizon stability","cross-route hybridization"],"ai_summary_model":"thinkingmachines/Inkling-Small"},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"665c65f1987f055eff6c8ea5","avatarUrl":"/avatars/980de0ae1964971259e051764a784c55.svg","isPro":false,"fullname":"Frank W","user":"frankw132","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":0,"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2608/2608.23070.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

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

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

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

Add this paper to a collection to link it from this page.