PAWBench: How Far Are We from Probabilistically Aligned World Modeling?
[Widzenie komputerowe] Autorzy: Yuandong Pu, Le Zhuo, Sayak Paul i 11 więcej
Recent video generation models are increasingly framed as world models. Many physical processes can unfold in more than one valid way. Therefore, a world model should reproduce not only a plausible trajectory, but also the distribution of possible behaviors under the same initial observation and action. We call this distribution-level requirement probabilistic alignment. However, existing evaluations largely assess individual-video plausibility and do not test whether repeated generations recover the correct distribution. This raises a central question: how far are current video generators from probabilistically aligned world modeling? To answer it, we formalize probabilistic alignment as a distributional criterion for world models and introduce PAWBench, a benchmark for evaluating video generators as stochastic samplers of world dynamics. We further introduce PAWEval, an outcome-level …
- Kategoria arxiv
- cs.CV
- Kategoria
- Widzenie komputerowe