EV-WM: Event-Verified World Models for Long-Horizon Robotic Manipulation

arXiv 2026 53 method, application

TLDR

EV-WM introduces predicate-grounded verification for world-model planning in long-horizon robotic manipulation, improving interpretability and task alignment.

Reasoning

Strengths: novel verification framework using event states and multiple scoring terms, tested on diverse manipulation tasks. Weaknesses: lacks explicit real-world validation and quantitative comparisons; reliance on pretrained features may limit generalization.

Read-first score

Read-first score 53, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 43.

Recency 6%

100

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

Citation impact 18%

86.8

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.86841732

Topical relevance 29%

61.4

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

Methodology quality 18%

Screens visible abstract and analysis fields for experiment, dataset, baseline, metric, and limitation evidence. markers=none

Reproducibility 18%

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

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: 378.

Keyword Scores

world model

world dynamics prediction

generative world model

model-based reinforcement learning world model

world simulator

interactive world model

video world model

Deep Analysis

Innovations

Predicate-grounded verification framework for world-model planning (EV-WM)
Rolls out candidate futures in pretrained visual-feature space, decodes into structured event states, and scores using task-progress, semantic-consistency, physical-feasibility, and uncertainty terms
Verifier guides sampling-based planning, gates candidate actions, and selects among PPO-generated proposals in contact-sensitive settings
Makes feature-space world-model planning more interpretable and better aligned with task progress

Methodology

EV-WM uses pretrained-feature world models to roll out candidate futures, decodes them into structured event states, and scores them using task-progress, semantic-consistency, physical-feasibility, and uncertainty terms. The verifier guides sampling-based planning, gates candidate actions, and selects among PPO-generated proposals in contact-sensitive settings such as the LIBERO wine-rack task.

Key Results

Across navigation, deformable-object, wall-constrained, and language-described manipulation studies, EV-WM demonstrates that predicate-grounded verification can make feature-space world-model planning more interpretable and better aligned with task progress.

Resources

Paper