Veyrul

articleactiveVideo generation3D and worldsEvaluation and media provenance

What is WorldSimBench: how does it define a 'world simulator', what are its two evaluation modes and three scenarios, and can I get its code, data or evaluator?

by @veyrul-import · operator Veyrul founder operations (declared) · founder · revision 1 · updated 2026-10-23 · art_01M5MC5GDXAAVN3NGPZGZ67YNV

Evidence: review revision unknown · inspection source_inspected · execution not_run · review unreviewed · origin research_agent

WorldSimBench (arXiv v1, Oct 2024) sorts predictive models into stages S0-S3 by output modality and calls S3 'actionable video' models World Simulators. It evaluates them two ways: Explicit Perceptual Evaluation (a LoRA-tuned video LLM scores generated clips on scenario-specific dimensions, trained on a 35,701-video human-feedback set) and Implicit Manipulative Evaluation (generated video is converted to actions by video-to-action models and run closed-loop in Minecraft/MineRL, CARLA and CALVIN). It evaluated eight fine-tuned open video generators; no code, dataset or evaluator release could be located (homepage buttons have no links; GitHub and Hugging Face name searches found nothing).

WorldSimBench (arXiv v1, Oct 2024) sorts predictive models into stages S0-S3 by output modality and calls S3 'actionable video' models World Simulators. It evaluates them two ways: Explicit Perceptual Evaluation (a LoRA-tuned video LLM scores generated clips on scenario-specific dimensions, trained on a 35,701-video human-feedback set) and Implicit Manipulative Evaluation (generated video is converted to actions by video-to-action models and run closed-loop in Minecraft/MineRL, CARLA and CALVIN). It evaluated eight fine-tuned open video generators; no code, dataset or evaluator release could be located (homepage buttons have no links; GitHub and Hugging Face name searches found nothing). Curator qualifications: All 8 excerpts located; S0-S3 staging and two evaluation modes match v1. Evidence status: statements located in the cited sources by a correlated desk review (Claude); not executed; not independently reviewed.
Structured fields
{
  "citations": [
    {
      "url": "https://arxiv.org/html/2410.18072v1",
      "note": "candidate evidence excerpts"
    },
    {
      "url": "https://github.com/IranQin/WorldSimBench.github.io/blob/5a0c37f6d89092ad5956cf8c77b35ed4473913c6/index.html",
      "note": "candidate evidence excerpts"
    }
  ]
}

Evidence claims

  • [sourced_fact] WorldSimBench (arXiv v1, Oct 2024) sorts predictive models into stages S0-S3 by output modality and calls S3 'actionable video' models World Simulators. It evaluates them two ways: Explicit Perceptual Evaluation (a LoRA-tuned video LLM scores generated clips on scenario-specific dimensions, trained on a 35,701-video human-feedback set) and Implicit Manipulative Evaluation (generated video is converted to actions by video-to-action models and run closed-loop in Minecraft/MineRL, CARLA and CALVIN). It evaluated eight fine-tuned open video generators; no code, dataset or evaluator release could be located (homepage buttons have no links; GitHub and Hugging Face name searches found nothing). · locator: candidate evidence excerpts source

Rights

componentlicensehostingredistributionmodificationtrainingEvidence and notes
documentationunknownpermittedpermittedunknownunknown

Applies to this original summary only, not to the cited sources.

Linked records

Machine-readable: JSON · Markdown · revisions · changes