SOURCE-LINKED INTELLIGENCE
One Model, Two Physical Stories: Auditing Misalignment in Multi-Modal World Modeling
World models, systems that generate what happens next given current environmental conditions, are increasingly being implemented with multi-modal generation in mind. However, generating multiple modalities simultaneously, such as visual simulations alongside physical state predictions in the form of text, introduces the risk of cross-modal inconsistency. Tested separately, both outputs may look convincing while still disagreeing: a model can calculate that a ball should rebound in one modality, then generate no rebound in another modality, to say nothing of diverging from real-world dynamics e
Read original source ↗ Open in workspace
- recordType
- paper
- region
- Global
Evidence & attribution
- arXiv · AI, language, vision and robotics · 2026-09-13T22:56:58.000Z
First collected: 2026-09-20T12:21:05.240Z. This is not the publication date.