Quantitative Video World Model Evaluation for Geometric-Consistency

arXiv 2026 55.5 benchmark

TLDR

Introduces PDI-Bench, a quantitative framework to evaluate geometric coherence in generated videos, revealing failure modes not captured by perceptual metrics.

Reasoning

The paper presents a novel quantitative evaluation framework for geometric consistency in generative video models, which is a strength. However, it relies on monocular reconstruction and point tracking, which may introduce errors, but the abstract does not discuss limitations.

Read-first score

Read-first score 55.5, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 35.

Recency 6%

100

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

Citation impact 18%

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.77997089

Methodology quality 18%

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

Topical relevance 29%

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=code,dataset,github

Citation velocity 12%

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

Field roles

FoundationFrontierBridgeMethodology anchor

Rank sensitivity

Stability: volatile; rank range: 331.

Keyword Scores

video world model

generative world model

world model

world dynamics prediction

world simulator

interactive world model

model-based reinforcement learning world model

Deep Analysis

Innovations

PDI-Bench: a quantitative framework for evaluating geometric coherence in generated videos using projective-geometry residuals
PDI-Dataset: a dataset covering diverse scenarios designed to stress geometric constraints
Identification of geometry-specific failure modes not captured by common perceptual metrics

Methodology

Given a generated clip, object-centric observations are obtained via segmentation and point tracking (e.g., SAM 2, MegaSaM, and CoTracker3), lifted to 3D world-space coordinates via monocular reconstruction, and a set of projective-geometry residuals is computed capturing three failure dimensions: scale-depth alignment, 3D motion consistency, and 3D structural rigidity.

Key Results

Across state-of-the-art video generators, PDI reveals consistent geometry-specific failure modes that are not captured by common perceptual metrics, providing a diagnostic signal for progress toward physically grounded video generation and physical world models.

Resources

Paper