Hallucination in World Models is Predictable and Preventable

arXiv 2026 53 method, benchmark

TLDR

Hallucination in world models stems from low data coverage; detectable via three signals and preventable with coverage-aware sampling and curiosity-driven finetuning.

Reasoning

The paper introduces a large dataset (MMBench2) and identifies three hallucination modes, offering both detection signals and mitigation strategies. Strengths include empirical grounding and practical finetuning with few trajectories; weaknesses are the reliance on a single model size (350M) and limited detail on generalizability beyond the dataset.

Read-first score

Read-first score 53, weighted from topical fit, citation, graph, method, reproducibility, and recency signals. Original total remains 55.

Recency 5%

100

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

Citation impact 15%

95.6

Uses OpenAlex-shaped citation metadata as a bibliometric attention signal, separate from paper quality. citation_normalized_percentile=0.95643432

Topical relevance 25%

78.6

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 15%

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

Reproducibility 15%

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

Graph prestige 15%

5.7

PageRank-style prestige over the local citation graph built from available OpenAlex references. local_pagerank=0.0016

Citation velocity 10%

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

Field roles

FoundationFrontier

Rank sensitivity

Stability: volatile; rank range: 498.

Keyword Scores

world model

generative world model

interactive world model

world dynamics prediction

video world model

model-based reinforcement learning world model

world simulator

Deep Analysis

Innovations

Identification of three distinct hallucination modes: perceptual, action-marginalized, and scene-diverging
Development of three signals that accurately predict where the world model will fail
Coverage-aware sampling technique to close coverage gaps at training time
Hallucination predictors used as curiosity rewards for targeted data collection, enabling data-efficient finetuning with as few as 50 real environment trajectories
Introduction of MMBench2, a 427-hour, 210-task dataset for visual world modeling with ground-truth actions, rewards, and live simulators

Methodology

The authors train a 350M-parameter world model on the MMBench2 dataset. They identify three hallucination modes and develop three predictive signals. To mitigate, they use coverage-aware sampling during training and hallucination predictors as curiosity rewards for online data collection, enabling finetuning on unseen environments with few trajectories.

Key Results

Hallucination in world models is predictable and preventable; the same data-centric signals used for detection can also guide mitigation, allowing adaptation to unseen environments with as few as 50 real trajectories.

Resources

Paper