Awesome World Model Hub Papers · Datasets · Projects
← Back to papers

A Step Toward World Models: A Survey on Robotic Manipulation

arXiv 25.11 2025 57.6 survey

TLDR

A survey examining world model capabilities in robotic manipulation, analyzing perception, prediction, and control to define core components.

Reasoning

Strengths: Provides a structured analysis of world model concepts in robotic manipulation, clarifying ambiguous definitions. Weaknesses: As a survey, it lacks novel experiments or real-world validation; the scope is limited to manipulation, potentially missing broader applications.

Read-first score

Read-first score 57.6, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 35.

Recency 8%
86.7

Uses a gentle age decay so recent papers surface without erasing older foundations. 2025

Methodology quality 25%
80

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

Topical relevance 42%
50

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 25%
38

Screens links and visible text for paper, code, dataset, artifact, and repository signals. pdf=True; code=False; dataset=False; markers=code

Field roles

FrontierMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 309.

Keyword Scores

world model
10
world dynamics prediction
8
model-based reinforcement learning world model
5
world simulator
4
generative world model
3
interactive world model
3
video world model
2

Deep Analysis

Innovations

  • Goes beyond prescribing a fixed definition of world models, instead examining methods that exhibit core capabilities
  • Analyzes world model approaches across perception, prediction, and control in robotic manipulation
  • Distills core components, capabilities, and functions that a fully realized world model should possess

Methodology

This survey reviews methods in robotic manipulation that exhibit core capabilities of world models, analyzing their roles in perception, prediction, and control, and identifying key challenges and solutions. It does not limit itself to methods explicitly labeled as world models, but rather examines approaches that demonstrate world model-like behaviors.

Key Results

The survey identifies key challenges and solutions in robotic manipulation world models, and distills the core components, capabilities, and functions that a fully realized world model should possess.

Limitations

  • Scope is limited to robotic manipulation, not covering other domains like navigation or decision-making
  • Does not provide a fixed definition of world models, leaving ambiguity in scope and architecture
  • As a survey, it does not present new experimental results or validate proposed components empirically

Tags