paper
Hallucination in World Models is Predictable and Preventable
paperactiveprovisional
hallucination-in-world-models-is-predictable-and-preventable-4df7c11b·1 events·first seen 4d agoAliases: Hallucination in World Models is Predictable and Preventable
Co-occurring entities
More like this (12)
Vision-Default, Prior-Override: Causal Mechanisms of Perception-Knowledge Conflict in Vision-Language ModelsLooped World Modelsstable-worldmodelPredicting Future Behaviors in Reasoning Models Enables Better SteeringHallucinations Leaderboardhallucination (LLM)world modelphysical world modelRecalling Too Well: Sycophancy Evaluation and Mitigation in Memory-Augmented ModelsWhen the Chain of Thought Knows Better: Failure Modes in Multi-Turn Reasoning ModelsLanguage Modeling LossA Causal Model of Theory of Mind in Conflict for Artificial Intelligence
Recent events (1)
MMBench2 paper: hallucination in world models is predictable and preventable via coverage signals
Researchers introduce MMBench2, a 427-hour, 210-task dataset for visual world modeling, and train a 350M-parameter world model to study hallucination in generative world models. The paper identifies three distinct hallucination modes (perceptual, action-marginalized, scene-diverging) and develops lightweight signals that predict where models will fail. A coverage-aware sampling technique and curiosity-reward-based data collection enable efficient finetuning to unseen environments with as few as 50 real trajectories. The central finding is that world model hallucination is fundamentally a data coverage problem, with the same signals serving both detection and mitigation.