A new 23-dimension benchmark finds that even the best VLMs score near random on motion trajectory, temporal extension, and several prediction tasks, far below humans, suggesting weak internal world models.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Do Vision-Language Models Have Internal World Models? Towards an Atomic Evaluation
A new 23-dimension benchmark finds that even the best VLMs score near random on motion trajectory, temporal extension, and several prediction tasks, far below humans, suggesting weak internal world models.