OptiWorld: Optimal Control for Video World Generation under Physical Constraints

arXiv 2026 54.4 method

TLDR

OptiWorld integrates optimal control with video generation to produce physically consistent, goal-driven videos at inference time.

Reasoning

The paper presents a novel framework combining optimal control and video generation, addressing physical consistency and trajectory optimization. Strengths include a clear methodology and multiple task demonstrations, but weaknesses are the lack of real-world experimental validation and reliance on synthetic or unspecified benchmarks.

Read-first score

Read-first score 54.4, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 40.

Recency 6%

100

Uses a gentle age decay so recent papers surface without erasing older foundations. 2026

Citation impact 18%

89.8

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.89760603

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=metric,result

Topical relevance 29%

57.1

Uses existing LLM keyword relevance scores normalized to 0-100. world model,world simulator,generative world model,interactive world model,video world model,world dynamics prediction,model-based reinforcement learning world model

Reproducibility 18%

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=none

Citation velocity 12%

Citation velocity estimates citations per publication-year to reduce old-paper bias. velocity=0.00

Field roles

FoundationFrontierBridge

Rank sensitivity

Stability: volatile; rank range: 377.

Keyword Scores

world model

video world model

generative world model

world dynamics prediction

world simulator

model-based reinforcement learning world model

interactive world model

Deep Analysis

Innovations

Bringing classical optimal control into video generation at inference time
Extracting a compact, task-relevant world state for planning
Formulating planning as a geometric problem on a continuous manifold that converts 3D geometry and physical constraints into a unified planning geometry
Application to multiple tasks: goal-conditioned image-to-video generation, video dynamics editing, and counterfactual generation

Methodology

OptiWorld first extracts a compact, task-relevant world state from the input, then plans an optimal trajectory under physical constraints by formulating planning as a geometric problem on a continuous manifold, and finally renders the video conditioned on this trajectory. The planning converts 3D geometry and task-dependent physical constraints into a unified planning geometry.

Key Results

OptiWorld generates videos with preferable dynamics under physical constraints, demonstrating strong potential in goal-conditioned image-to-video generation, video dynamics editing, and counterfactual generation.

Resources

Paper